A Dangerous Concoction: Pharmaceutical Marketing

Loyola University Chicago, School of Law Loyola University Chicago, School of Law

LAW eCommons LAW eCommons

Faculty Publications & Other Works

2019

A Dangerous Concoction: Pharmaceutical Marketing, Cognitive A Dangerous Concoction: Pharmaceutical Marketing, Cognitive

Biases, and First Amendment Overprotection Biases, and First Amendment Overprotection

Cynthia M. Ho

Follow this and additional works at: https://lawecommons.luc.edu/facpubs

Part of the First Amendment Commons, and the Health Law and Policy Commons

https://lawecommons.luc.edu/

https://lawecommons.luc.edu/facpubs

https://lawecommons.luc.edu/facpubs?utm_source=lawecommons.luc.edu%2Ffacpubs%2F635&utm_medium=PDF&utm_campaign=PDFCoverPages

http://network.bepress.com/hgg/discipline/1115?utm_source=lawecommons.luc.edu%2Ffacpubs%2F635&utm_medium=PDF&utm_campaign=PDFCoverPages

http://network.bepress.com/hgg/discipline/901?utm_source=lawecommons.luc.edu%2Ffacpubs%2F635&utm_medium=PDF&utm_campaign=PDFCoverPages

A DANGEROUS CONCOCTION: PHARMACEUTICALMARKETING, COGNITIVE BIASES, AND FIRST AMENDMENT

OVERPROTECTION

CYNTHIA M. Ho*

This Article argues that pharmaceutical marketing to doctors should be morecritically evaluated and entitled to less First Amendment protection, contrary to atrend dating back to the Supreme Court's 2011 decision in Sorrell. In particular, theArticle argues that more information to doctors in the form of pharmaceuticalmarketing does not necessarily result in better patient outcomes. The Article adds asignificant critique based on the existence and impact of cognitive bias literature thathas thus far not been recognized in this area. If courts fully embrace thisunderstanding, they should recognize that the government, through the Food andDrug Administration, has a right to limit statements that may encourage doctors toprescribe unapproved uses of drugs with potentiallyfatal consequences.

This Article reveals that recent expansion of First Amendment jurisprudence isbased on key cognitive biases and assumptions. First, courts, and even some doctorsthemselves, improperly assume that doctors are adequately sophisticated, such thatdoctors are protectedfrom self interested marketing, which this Article demonstratesas inconsistent with reality. Second, current case law assumes that the availabilityofmore information necessarilypromotes better decisions so long as it is notpatentlyfalse, a proposition that this Article shows is especially unfounded in the uniquemarket ofprescription drugs. Importantly, such assumptions can have critical healthconsequences since they promote uses of drugs for which there is often inadequatescientific basis and serious health consequences.

Finally, this Article builds upon the revealed cognitive biases to suggestempirically-informed changes to cabin the expansion of First Amendment protectionof pharmaceutical marketing as well as broader structural reform. This Articleproposes to treat potentially misleading information differently than entirely truthfulspeech, thus giving states greater discretion to regulate potentially misleadinginformation. In addition, this Article proposes that the burden ofproof in such casesshould be reversed, so that courts will no longer consider disclaimers as a truealternative to speech restriction without proof that companies will actually promotemore informed decisions. The Article also suggests structural changes to medicaleducation, drug development, and marketing informed by the cognitive biasesrevealed here.

* Clifford E. Vickrey Research Professor, Loyola University Chicago School of Law.

This Article benefitted from comments by participants at ASMLE 2017, the Wiet 2017 LifeSciences Conference, and a 2018 Loyola faculty workshop. Thanks to Leah Chan Grinvald,Dmitry Karshtedt, Joan Krause, Christopher Robertson, and Sarah Waldeck for comments onearlier drafts. Extra special thanks go to Jordan Paradise and Nadia Sawiki for providingincredibly helpful feedback and support on multiple drafts. This Article also benefited fromthe research and editorial assistance of Teresa Gaglione, Austin Holler, Sarah Johnson, TedMahan, Monika Malek, Catie Eubanks, Kate Finch, and Cameron Wooley.

INDIANA LAW JOURNAL

IN TR O D U C T IO N ...................................................................................................... 7 7 5

1. B A C K G R O U N D ................................................................................................ 7 8 1

A. DEVELOPMENT AND APPROVAL OF PRESCRIPTION DRUGS ................... 781B. M ARKETING OF PRESCRIPTION DRUGS ................................................. 785

C. FDA REGULATION OF OFF-LABEL USES .............................................. 792D. INTRODUCTION TO COMMERCIAL SPEECH LAW AND POLICY ............... 797

II. REVEALING AND DEBUNKING SCHEMAS THAT SUPPORT FIRST

A M END M ENT FALLACIES ............................................................................... 801

A . INTRODUCTION TO SCHEM AS ................................................................ 8021. SCHEMAS SHAPE INFORMATION PROCESSING ............................. 8032. SCHEMAS ARE REINFORCED BY OTHER COGNITIVE BIASES ........ 804

B. THE SOPHISTICATED DOCTOR SCHEMA ................................................ 8071. WHAT IS THE SOPHISTICATED DOCTOR SCHEMA? ......... . .. .. .. .. .. .. 8072. DEBUNKING THE SOPHISTICATED DOCTOR SCHEMA ................... 813

C. THE M ORE INFORMATION SCHEMA ...................................................... 8181. WHAT IS THE MORE INFORMATION SCHEMA? .............. .. .. .. .. .. .. .. 819

2. EVIDENCE DEBUNKING "MORE INFORMATION IS

B ETTER" SCHEM A ....................................................................... 820

III. IM PL IC A T IO N S ................................................................................................ 82 6A. RETROSPECTIVE VIEW ON FIRST AMENDMENT EXPANSION

INFORMED BY COGNITIVE BIAS ............................................................ 8261. RELEVANT CASE LAW EXPANDING COMMERCIAL

SPEECH FOR OFF-LABEL PROMOTION ......................................... 8272. EXAMINING KEY CASES THROUGH THE LENS OF THE

RE VEALED SCHEM AS .................................................................. 828

I. FALSE OR INHERENTLY MISLEADING? .............. . .. .. .. .. .. .. .. .. 829II. SUBSTANTIAL GOVERNMENT INTEREST? ............ .. .. .. .. .. .. .. 831III. DIRECT ADVANCEMENT OF INTEREST? ............ . .. .. .. .. .. .. .. .. 832IV. N ARROW LY TAILORED ...................................................... 833

3. SCHEMAS BAKED INTO PRECEDENT RESULT IN FDAC AP ITU LA TIO N ............................................................................ 835

B. THE MISMATCH BETWEEN FIRST AMENDMENT POLICY AND

PHARMACEUTICAL M ARKETING ........................................................... 8361. POLICY SIMILARITIES BETWEEN INHERENTLY AND

POTENTIALLY M ISLEADING SPEECH ........................................... 8362. POTENTIALLY MISLEADING PHARMACEUTICAL MARKETING

IS UNLIKELY TO BE CORRECTED, WITH SIGNIFICANT PUBLIC

HARM , BUT LIM ITED BENEFIT ..................................................... 838IV. AN INFORMED APPROACH TO DRUG ADVERTISING AND DEVELOPMENT ....... 839

A. ALIGNING FIRST AMENDMENT LAW AND POLICY WITH REALITY ........ 8391. AN INFORMED TREATMENT OF POTENTIALLY MISLEADING

INFO RM A TIO N ............................................................................. 8402. A MODIFIED BURDEN ON DISCLAIMERS ...................................... 8423. MORE JUDICIAL DEFERENCE TO FDA EVALUATION ................... 8434. REEVALUATING FDA GUIDANCE ................................................ 844

B. SUGGESTIONS FOR STRUCTURAL REFORMS .......................................... 8451. INCREASE A W ARENESS ............................................................... 845

[Vol. 94:773

A DANGEROUS CONCOCTION

2. BEYOND CONFLICTS OF INTEREST - RECONSIDERING DRUG

SAM PLES AND G IFTS ................................................................... 8483. OTHER MECHANISMS TO MINIMIZE PUBLIC HARM FROM

S CH E M A S .................................................................................... 8 50

4. TOWARDS MORE INDEPENDENT AND ACCESSIBLE

IN FO R M A TIO N ............................................................................. 85 1

5. ADDITIONAL SUPPORT FOR RECONSIDERING THE DRUG

D EVELOPMENT PROCESS ............................................................. 852

C O N C LU SIO N ......................................................................................................... 8 54

INTRODUCTION

Is more information always better? First Amendment commercial speechjurisprudence takes this as a given.1 However, when information is only availablefrom a self-interested and marketing-savvy pharmaceutical company, moreinformation may simply lead to more misinformation. Notably, doctors are alsomisled. This can result in public health harms when companies are promotingunapproved uses of prescription drugs that the Food and Drug Administration (FDA)has approved for other purposes-commonly referred to as "off-label" uses.Contrary to judicial presumptions, as well as the presumptions of some doctors andscholars, doctors are not sophisticated enough to always discern what is true versusmisleading information. Doctors are susceptible to the same largely unconsciouscognitive biases as all individuals;' this means that they operate on "schemas"(mental presumptions) that impact how they interpret marketing information.3 Courtsalso rely on schemas about how doctors interact with marketing. These schemas havecontributed to a First Amendment jurisprudence that has serious consequences forpublic health because it fails to account for how doctors actually interact withmarketing of off-label uses, and such uses are associated with adverse healthconsequences.

This Article argues that pharmaceutical marketing, especially regarding "off-label" uses, should be more critically evaluated and entitled to less First Amendmentprotection-contrary to recent court trends, beginning with the 2011 Supreme Courtcase Sorrell v. IMS Health Inc.4 In other words, this Article is taking a new approachto address court cases that a number of scholars have criticized as unduly expandingthe scope of First Amendment protection for pharmaceutical marketing with negative

1. E.g., Peel v. Attorney Registration & Disciplinary Comn'n of Ill., 496 U.S. 91, 110(1990); In re R.M.J., 455 U.S. 191, 206 n.20 (1982); Bates v. Arizona, 433 U.S. 350, 374-75(1977) (rejecting suggestion that public is better off with incomplete, but correct, informationversus potentially misleading information in the context of attorney advertising).

2. See infra Section II.A.

3. The term "schema" is used here to refer to an organizing mental framework forprocessing information. See MARTHA AUGOUSTINOS, lAIN WALKER & NGAIRE DONAGHUE,

SOCIAL COGNITION: AN INTEGRATED INTRODUCTION 68-69 (3d ed. 2014).

4. 564 U.S. 552 (2011); see United States v. Caronia, 703 F.3d 149 (2d Cir. 2012);Amarin Pharma, Inc. v. FDA, 119 F. Supp. 3d 196 (S.D.N.Y. 2015); see also infra SectionIII.A. 1 (explaining expansion of commercial free speech through these cases).

2019]

INDIANA LAW JOURNAL

policy repercussions for public health.5 Not only are many off-label uses medicallyunsupported, but permitting broader promotion of such uses undermines incentivesfor companies to scientifically study those uses.6 Whereas prior articles have tendedto focus on how to adapt to the new law7 or advocate rejecting the existing law basedprimarily on policy grounds,8 this Article uses cognitive bias literature to explain

5. E.g., Stephanie M. Greene, After Caronia: First Amendment Concerns in Off-Label

Protection, 51 SAN DINGO L. REV. 645 (2014); Loren Jacobson, Don't Fix WhatAin 't Broken

Off-Label Marketing, the FDA's Regulatory Regime, and the First Amendment, 5 EMORY

CORP. Gov & ACCONTABILITY REV. 19 (2018); Aaron S. Kesselheim, Michelle M. Mello &

Jerry Avorn, FDA Regulation of Off-Label Drug Promotion Under Attack, 309 J. AM. MED.

ASS'N 445 (2013); David Orentlicher, Off-Label Drug Marketing, the First Amendment, andFederalism, 50 WASH. U. J.L. & POL'Y 89 (2016); Joshua M. Sharfstein, Public Health and

the First Amendment, 93 MILBANK Q. 459 (2015); Patricia J. Zettler, The IndirectConsequences of Expanded Off-Label Promotion, 78 OH. ST. L.J. 1053 (2017). Of course,there are also some scholars that embrace the judicial expansion of First Amendment rights.E.g., Coleen Klasmeier & Martin H. Redish, Off-Label Prescription Advertising, the FDA and

the First Amendment A Study in the Values of Commercial Speech Protection, 37 AM. J.L. &MED. 315 (2011); Lars Noah, The Little Agency That Could (Act with Indifference toConstitutional and Statutory Strictures), 93 CORNELL L. REV. 901 (2008); Rodney A. Smolla,Off-Label Drug Advertising and the First Amendment, 50 WAKE FOREST L. REV. 81 (2015). Inaddition, others argue for further expansion. E.g., Jeffrey Chasnow & Geoffrey Levitt, Off-Label Communications: The Prodigal Returns, 73 FOOD & DRUG L.J. 257 (2018); ChristinaSandefur, The FDA's Approach to Off-Label Communications: Restricting Free Speech inMedicine?, REG. TRANSPARENCY PROJECT (May 10, 2018), https://regproject.org/paper/fdas-

approach-off-label-communications-restricting-free-speech-medicine[https://perma.cc/FR5B-29CQ]; see also Carl Wiersum, No Longer Business as Usual: FDA

Exceptionalism, Commerical Speech, and the First Amendment, 73 FOOD & DRUG L.J. 486(2018) (arguing FDA rules regarding off-label promotion as well as claims for dietarysupplements and graphic ciagarette warnings are inconsistent with First Amendment).

6. E.g., Zettler, supra note 5; Memorandum from the Food & Drug Admin. on Pub.Health Interests and First Amendment Considerations Related to MFR. Commc'ns Regarding

Unapproved Uses of Approved or Cleared Med. Prods. 4-5 (Jan. 2017) [hereinafter FDAMemorandum].

7. E.g., Constance E. Bagley, Joshua Mitts & Richard J. Tinsley, Snake Oil Salesmen orPurveyors of Knowledge: Off-Label Promotions and the Commercial Speech Doctrine, 23CORNELL J.L. & PUB. POL'Y 337 (2013) (providing suggestions for FDA regulation afterCaronia that arguably meet the stricter heightened scrutiny announced in Sorrell); Aaron S.Kesselheim & Michelle M. Mello, Prospects for Regulation of Off-Label Drug Promotion in

an Era of Expanding Commercial Speech Protection, 92 N. C. L. REv. 1539 (2014) (providing

suggestions for future FDA action since Caronia).8. E.g., Orentlicher, supra note 5,passim (arguing that courts should reject the reasoning

of Caronia); Christopher Robertson, The Tip of the Iceberg: A First Amendment Right toPromote Drugs Off-Label, 78 OH. ST. L.J. 1019 (2017) (challenging arguments in favor of off-label marketing and noting that if these are adopted, all premarket approval of drugs are at risk

and also undermine other statutory regimes beyond the FDA); Christopher Robertson & AaronS. Kesselheim, Regulating Off-Label Promotion A Critical Test, 375 NEW ENG. J. MED.

2313, 2314 (2016) (arguing that courts should reject Caronia on its merits and that the FDA,not courts, should be evaluating whether corporate claims are valid); Christopher Robertson

& Victor Laurion, Tip of the Iceberg II: How the Intended- Uses Principle Procedure MedicalKnowledge and Protects Liberty, 11 N.Y.U. J.L. & LIBERTY 770 (2017) (explaining why

[Vol. 94:773


why recent cases actually fail to achieve key First Amendment goals. This Articlefurther provides an empirically supported argument against expansion of FirstAmendment law for off-label promotion.

Marketing of off-label uses is a unique area where First Amendment commerciallaw concepts may not work well because of an essentially imperfect market. As willbe explained, commercial law jurisprudence recognizes commercial speech asvaluable to listeners to the extent it provides information. Even though the lawrealizes some information is potentially misleading, it errs on the side of permittingsuch information on the assumption that the marketplace of ideas can providecompeting information.9 However, there is no functioning marketplace of ideas withnew drugs. Drugs are developed and sold initially only by self-interested companieswho must provide clinical data to the FDA, but not to the public. 10 Manufacturers ofcompeting drugs for the same condition do not contribute to a functioningmarketplace of ideas because companies tend to only promote attributes of their owndrug, but not in comparison to others; this is likely because regulations do not requirea drug be superior to existing treatments. 11 Accordingly, the only objective source ofinformation would be from independent scientists unassociated with self-interestedcompanies seeking approval. However, it takes years before independent scientistscan evaluate already approved drugs. Sometimes independent scientists will findimportant caveats to the initial assertions. 12 This is not entirely surprising since thedata submitted to the FDA is developed by the self-interested company.13 However,

existing FDA regulation are important to produce information and that it functions similar toother regulations). There is also substantial scholarship questioning expansion of FirstAmendment commercial speech, including Sorrell. E.g., Julie E. Cohen, The Zombie FirstAmendment, 56 WM. & MARY L. REV. 1119 (2015); Amanda Shanor, The New Lochner, 133Wis. L. REV. 133 (2016).

9. See, e.g., supra note 1.10. See infra Section I.A.11. E.g., infra note 33; see also Richard A. Friedman, What Drug Ads Don't Say, N.Y.

TIMES (April 23, 2016), https://www.nytimes.com/2016/04/24/opinion/sunday/what-drug-ads-dont-say.html [https://perma.cc/5MZP-UVEZ] (noting that most "new" drugs areunlikely to "substantially outperform" existing drugs). Even when companies tout a feature oftheir drug, it may misleadingly suggest that this is an improvement when in fact othercompanies simply did not seek similar approval. E.g., CYNTHIA M. Ho, ACCESS TO MEDICINE

IN THE GLOBAL ECONOMY: INTERNATIONAL AGREEMENTS ON PATENTS AND RELATED RIGHTS

31 (2011) (explaining that Zyrtec was marketed as the first approved allergy drug for indoorand outdoor allergies even though other allergy drugs are effective for this purpose as well).In addition, companies may intentionally not conduct studies in comparison to other drugs forfear that their drug may perform suboptimally. See, e.g., Ron Winslow, For Bristol-Myers,Challenging Pfizer Was a Big Mistake, WALL ST. J. (Mar. 9, 2004, 12:01 AM),https ://www.wsj.com/articles/SB 107876021684949151 [https://perma.cc/X76M-V2HH](noting that companies rarely do such studies in the context of discussing Bristol MyersSquibb's failed attempt to establish that its cholesterol-lowering drug compared well toPfizer's Lipitor).

12. See infra Section I.A.13. Some have suggested that given public interests, it would be better for the government

to pay for clinical trials and create independence. E.g., STAN FINKELSTEIN & PETER TEMIN,

REASONABLE RX: SOLVING THE DRUG PRICE CRISIS (2008); Tracy R. Lewis, Jerome H.

2019]

INDIANA LAW JOURNAL

before independent scientists can do their own studies, corporate marketing lacks anycounterbalance. Moreover, although the claims are allegedly based on scientificfacts, it is nonetheless possible to selectively present facts. Even if doctors are awareof the financial incentives to misrepresent, given the lack of competing information,they may nonetheless be influenced by these "facts." This is especially true sincecognitive bias literature shows that repeated facts are often presumed true. 14

The time is ripe to address the existence and impact of largely subconsciouscognitive biases to First Amendment case law. Other areas of the law and regulationare increasingly recognizing that subconscious cognitive biases should be given morecredence;i" for example, the Obama Administration issued an Executive Order to useinsights from behavioral science research to direct federal policies for betteroutcomes.16 Although a few scholars have recognized that modem marketing is oftenintended to target subconscious biases, they have not addressed the unique situationof alleged "facts" that cannot be verified and instead focused on general marketing,or information presumed to have no factual content based on images, packagingcolor, or product placement. 17 In contrast, no one has yet addressed the existence andimpact of cognitive biases on allegedly factual statements concerning off-label drug

Reichman & Anthony D. So, The Case for Public Funding and Public Oversight of ClinicalTrials, ECONOMISTS VOICE (Jan. 2007).

14. See infra notes 146-150.15. There is robust legal scholarship aimed at identifying and addressing unconscious,

typically racial biases in areas of policing, prosecuting, judicial decision-making, as well asEighth Amendment excessive punishment. E.g., Justin D. Levinson, Mark W. Bennett &Koichi Hioki, Judging Implicit Bias: A National Empirical Study of Judicial Stereotypes, 69FLA. L. REV. 63 (2017); Justin D. Levinson & Robert J. Smith, Systemic Implicit Bias, 126YALE L.J. F. 406 (2017); Justin D. Levinson & Danielle Young, Different Shades of Bias: SkinTone, Implicit Racial Bias, and Judgments ofAmbiguous Evidence, 112 W. VA. L. REv. 307(2010); L. Song Richardson, Police Efficiency and the Fourth Amendment, 87 IND. L.J. 1143,1144 (2012) (arguing that the legal test of"reasonable suspicion" is particularly susceptible toimplicit bias); Robert J. Smith & Justin D. Levinson, The Impact of Implicit Racial Bias onthe Exercise of Prosecutorial Discretion, 35 SEATTLE U. L. REv. 795 (2012).

16. Exec. Order No. 13,707, 3 C.F.R. § 13707 (2015).17. E.g., Micah L. Berman, Manipulative Marketing and the First Amendment, 103 GEO.

L.J. 497 (2015) (explaining how modem marketing is intended to take advantage of consumercognitive biases and proposing that "noninformational" marketing such as product placementand color should be granted less First Amendment protection in a manner consistent with theCentral Hudson test); Ralph S. Brown, Jr., Advertising and the Public Interest: LegalProtection of Trade Symbols, 57 YALE L. REv. 1165 (1948) (recognizing the powerfulinfluence of advertising with respect to trademarks); Yoav Hanner, Expressions WhichPreclude Rational Processing: The Case for Regulating Non-Informational Advertisements,27 WHITTIER L. REv. 435, 482 (2005); Christine Jolls, Debiasing Through Law and the FirstAmendment, 67 STAN. L. REV. 1411, 1419-36 (2015) (providing empirical data that compelleddisclosure of images in three situations may inform consumers contrary to current FirstAmendment case law that dismiss the ability of images to inform as not factual); see alsoTamara R. Piety, Merchants of Discontent An Exploration of the Psychology ofAdvertising,Addiction, and the Implications for Commercial Speech, 25 SEATTLE U. L. REv. 377, 407-21(2001) (discussing the psychology of advertising).

[Vol. 94:773


promotion whereas companies have for years been aptly relying on the existence ofcognitive biases of doctors with their marketing methods. 18

This Article argues that cognitive biases make doctors susceptible to thesealleged facts marketed by companies and that judges, too, have their own cognitivebiases that have thus far prevented them from seeing doctors' vulnerability tomanipulation. A better understanding of the cognitive biases at issue reveals that thecurrent trend towards greater First Amendment protection of commercial speech forpharmaceutical promotion is unlikely to achieve the traditional goal in FirstAmendment commercial speech jurisprudence of promoting informed decisions.Moreover, recognizing cognitive biases helps to establish why disclaimers areineffective, contrary to assumptions in First Amendment case law that suchdisclaimers are an easy alternative to speech restrictions.1 9

This Article complements existing knowledge and concern about pharmaceuticalmarketing, as well as broader commercial influences on the practice of medicine. Forexample, in recent years, scholars and policymakers have expressed concern thatdoctors may be unduly influenced by pharmaceutical marketing, gifts, industry-funded continuing medical education, and even "opinion leaders" with financial tiesto the industry.21 Some of this concern resulted in the Sunshine Act, which imposednew regulations to minimize financial conflicts of interest between doctors andindustry through disclosure of payments and items of certain value, but excluding

18. E.g., Sunita Sah & Adriane Fugh-Berman, Physicians Under the Influence: SocialPsychology and Industry Marketing Strategies, 41 J.L. MED. & ETHICS 665 (2013) (notingeffective use of social psychology principles by pharmaceutical marketing to which doctorsare generally unaware).

19. See infra Section II.C.20. E.g., CARL ELLIOTT, WHITE COAT, BLACK HAT: ADVENTURES ON THE DARK SIDE OF

MEDICINE (2010); INST. OF MED. OF THE NAT'L ACADS., CONFLICT OF INTEREST IN MEDICAL

RESEARCH, EDUCATION, AND PRACTICE (Bernard Lo & Marilyn J. Field eds., 2009); see alsoRosa Ahn, Alexandra Woodbridge, Ann Abraham, Susan Saba, Deborah Korenstein, ErinMadden, W. John Boscardin & Salomeh Keyhani, Financial Ties of Principal Investigatorsand Randomized Controlled Trial Outcomes: Cross Sectional Study, 356 BRIT. MED. J. 6(2017) (linking financial conflicts with publication bias); William Fleischman, ShantanuAgrawal, Marissa King, Arjun K. Venkatesh, Harlan M. Krumholz, Douglas McKee, DouglasBrown & Joseph S. Ross, Association Between Payments from Manufacturers of

Pharmaceuticals to Physicians and Regional Prescribing: Cross Sectional Ecological Study,354 BRIT. MED. J. 4-5, 7 (2016) (financial conflicts linked to prescribing more expensivedrugs); Ashley Wazana, Physicians and the Pharmaceutical Industry: Is a Gift Ever Just aGift?, 283 J. AM. MED. ASS'N. 373 (2000); Charles Ornstein, Mike Tigas & Ryann GrochowskiJones, Now There's Proof Docs Who Get Company Cash Tend to Prescribe More Brand-Name Meds, PROPUBLICA (Mar. 17, 2016, 5:00 AM), https://www.propublica.org/article/doctors-who-take-company-cash-tend-to-prescribe-more-brand-name-drugs [https://perma.cc/E83G-X4VY] (providing data concerning higher prescriptions of expensive drugs bydoctors who received more money as well as the fact that about seventy-five percent of doctorsin five common medical specialties received at least one payment in 2014). Even a single mealcan influence physician prescribing. E.g., Colette DeJong, Thomas Anguilar, Chien-WenTseng, Grace A. Lin, W. John Boscardin & R. Adams Dudley, Pharmaceutical Industry-Sponsored Meals and Physician Prescribing Patterns for Medicare Beneficiaries, 176 J. AM.MED. ASS'N 1114 (2016).

2019]

INDIANA LAW JOURNAL

others, such as drug samples.21 However, not only is disclosure an incompletesolution, but also the problem is not solely limited to conflict of interest. 22 After all,even doctors with no financial interest may have biases. This Article complementsrecognized problems with commercial influences on the practice of medicine butprovides a richer explanation than prior scholarship focused on conflicts of interest.Although some issues could be addressed through changes to malpractice law, thefocus of this Article is on addressing schemas in First Amendment law.2 3

This Article proceeds in four parts. Part I provides background to understand thekey issues. Part I begins with some fundamental concepts of how prescription drugsare developed and approved for sale. In addition, it discusses the unique nature of theprescription drug market, as well as how off-label uses are regulated. Finally, thisPart provides an overview of commercial speech law and policy.

Part II provides an introduction to schemas, as well as related cognitive biases asa backdrop to establishing two previously unrecognized schemas that have played akey role in promoting First Amendment fallacies concerning commercial speech.The first schema is that doctors are adequately sophisticated to evaluatepharmaceutical marketing of prescription drugs without being confused or misled(the "sophisticated doctor schema"). The second schema is that more information is

21. 42 U.S.C. § 1320a-7h (2012). Of course, there are still some outliers that claimconflict of interest is a mere narrative without adequate evidence of negative patient impacts.E.g., D. Barton, T. Stossel & L. Stell, After 20 Years, Industry Critics, Bury Skeptics, DespiteEmpirical Vacuum, 68 INT'L J. CLINICAL PRAC. 666 (2014).

22. See Sandra H. Johnson, Polluting Medical Judgment? False Assumptions in thePursuit of False Claims Regarding Off-Label Prescribing, 9 MINN. J.L. Sci. & TECH. 61, 63-64, 74-76 (2008) (considering conflicts of interest at best a partial accounting of off-labelprescribing and that doctors may be simply skeptical about clinical trials); Lisa Rosenbaum,Understanding Bias The Case for Careful Study, 372 NEW ENG. J. MED. 1959, 1960 (2015)(noting that conflict of interest disclosure may be interpreted by viewers as a reason to be moretrusting of the information because the disclosure is viewed as a sign of honesty). Moreover,even if disclosures of conflicts of interest provide some value, it seems that disclosures areincomplete since publications seem to rely on authors to voluntarily disclose and do notimpose penalties. E.g., Cole Wayant, Erick Turner, Chase Meyer, Philip Sinnett & MattVassar, Financial Conflicts of Interest Among Oncologist Authors of Reports of Clinical DrugTrials, J. Am. MED. Ass'N ONCOLOGY (2018) (noting that about a third of authors in a sampleof cancer trials did not report all payments); see also Charles Ornstein & Katie Thomas, TopCancer Researcher Fails to Disclose Corporate Financial Ties in Major Research Journals,N.Y. TiMEs (Sept. 8, 2018), https://www.nytimes.com/2018/09/08/health/jose-baselga-cancer-memorial-sloan-kettering.html [https://perma.cc/LD4Q-ESLU] (noting a doctor who failed to disclose conflicts of interest in dozensof research articles from a variety of prestigious journals and that most journals do not verifyaccuracy of information).

23. Although not the focus here, if doctors are less sophisticated in evaluating informationthan commonly believed, that could suggest changes to medical malpractice laws. Indeed,some have previously suggested that malpractice laws are ineffective in limiting off-label usethat could be dangerous. E.g., Phlip M. Rosoff & Doriane Lambelet Coleman, The Case forLegalRegulation of Physicians' Off-Label Prescribing, 86 NOTREDAMEL. REv. 649, 666-71(2011) (suggesting direct regulation of off-label use by limiting prescriptions by doctors forscientifically unsupported uses). However, even if medical malpractice might address someissues, that does not undermine the issue this Article addresses.

[Vol. 94:773


always desirable so long as it is not patently false, even if it is potentially misleading(the "more information schema"). Although each of these schemas has some basis inreality, current jurisprudence unduly relies on these schemas to an extent notsupported by empirical data. As will be explained, First Amendment case law hasthus far assumed that doctors will not be improperly swayed by "facts," given theirsophistication, contrary not only to cognitive bias literature, but growing recognitionthat doctors are vulnerable to marketing practices as recognized by the enactment ofthe Sunshine Act. In addition, although First Amendment jurisprudence regardingcommercial speech consistently promotes a robust marketplace of ideas, this iscontrary to literature about how all individuals incompletely process information, aswell as the fact that the pharmaceutical marketplace is an unusual one where there isgenerally no competing viewpoint.

Part III then considers the implications of the revealed schemas. Importantly, theschemas are not intended to suggest that all off-label marketing is problematic, orthat courts should permit an absolute ban on such promotion. This part considers keythemes in recent cases and also explains the mismatch between First Amendmentpolicy with the unique and imbalanced marketplace of patented pharmaceuticals.

Part IV provides concrete suggestions for how to combat the revealed schemas.This Part first suggests changes to the law to better align the law with reality,including modifications to commercial speech law relating to off-label use, as wellas related suggestions regarding judicial deference to FDA evaluation, andsuggestions for future FDA guidance. In addition, this Part suggests reforms thatwould help address the undue influence of schemas on the practice of medicinegenerally. Although these strategies would help address the revealed schemas in thecontext of off-label marketing of drugs, they have broader implications. Some of theproposed changes to the law impact how commercial speech regarding dietarysupplements, and especially disclaimers regarding such supplements are evaluated.In addition, some proposed structural changes, such as limiting drug samples wouldimpact all prescription drugs. The implications also extend to broader issues in thefield of medicine, such as the need for more independent data, including comparativeeffectiveness data, as well as questioning the current system of drug development.

I. BACKGROUND

A. Development and Approval ofPrescription Drugs

The development and approval of prescription drugs is important to understandsince it impacts both marketing and commercial speech implications. Importantly,although drugs can be considered public goods that benefit society, they aredeveloped by for-profit companies.24 Since companies have incentives to maximizeprofits, rather than focus on public health imperatives, regulation is important.Currently, before a new drug can be legally sold, its manufacturer must provideclinical data to the FDA that establishes that the drug is safe and effective for

24. Given this conflict of interest, some have suggested public funding should promotedrug development, and especially clinical trials needed to establish regulatory approval. E.g.,Lewis et al., supra note 13, at 1.

2019]

INDIANA LAW JOURNAL

intended use(s), in what is referred to as premarket approval.2" The current regulatoryscheme was established to avoid public health tragedies of an earlier era where drugswere not regulated at all, or were only regulated for safety, but not efficacy. 26

When the FDA evaluates whether a proposed new drug has adequate data toestablish safety and efficacy, it focuses on the drug's specific "indication(s)," i.e.use(s) for which the manufacturer seeks approval.27 Basically, the FDA isconsidering whether a drug's overall benefit versus risks for a specific use aresatisfactory. After all, a drug is unlikely to be effective and/or safe for all conditionsor all individuals.28 In addition, more risk may be tolerable for some conditions wherethere are no strong alternatives. The evaluation of drugs based on intended use isimportant to the FDA's mission in preventing unnecessary public harm.29 Inparticular, premarket approval for each intended use is essential to prevent acompany from obtaining approval for one use, but then promoting it for a differentunapproved use, without adequate scientific basis for the new use.31 In addition,approval based on specific uses encourages companies to conduct studies to establishwhether uses are supported.

While companies must provide substantial clinical data involving multiple phasesof human testing to the FDA that indicate the drug is safe and effective for itsproposed use(s), FDA approval does not completely guard against problems. First ofall, the self-interested company is designing studies to show that its drug is safe andeffective. In addition, although the company must submit the results of all clinicalstudies,31 the FDA may approve a drug despite equivocal clinical studies32 since the

25. 21 U.S.C. § 355 (2012).26. Drug Industry Antitrust Act of 1962: Hearing on H.R. 6245 Before the H. Comm. on

the Judiciary, 87th Cong. 67 (1962) ("[T]he physician is bombarded with seductiveadvertising which fails to tell the truth .... This often misleads him into prescribing a newdrug without adequate warning or information about its possible side effects and, indeed,without any solid clinical evidence that the drug is effective or is even as safe as the advertisersclaim."); Henry A. Waxman, A History of Adverse Drug Experiences: Congress Had AmpleEvidence to Support Restrictions on the Promotion of Prescription Drugs, 58 FOOD & DRUG

L.J. 299, 301-08 (2003).27. If approved, the drug "labeling," which includes not just the physical label on a drug,

but also anything distributed in connection with the drug, must reflect the approved indication.21 U.S.C. § 355(d); 21 C.F.R. § 202.1 (2017).

28. United States v. Rutherford, 442 U.S. 544, 555 (1979); see STEPHEN J. CECCOLI, PILL

POLITICS: DRUGS AND THE FDA 165-68 (2004).29. What We Do, FDA, https://www.fda.gov/aboutfda/whatwedo [https://perma.cc

/FW7W-SHMR]; see also FDA Memorandum, supra note 6 (noting that the FDA approvalbased on intended use was developed by Congress in response to public health tragedies).

30. S. REP. No. 87-1744 (1962).31. 21 C.F.R. § 314.50 (2018).32. For example, a FOIA request revealed that the six most widely prescribed

antidepressants approved between 1987 and 1999 had a combined total of forty-seven efficacystudies, and that more than half of these showed no significant difference between theapproved drug and placebo. Irving Kirsch, Thomas J. Moore, Alan Scoboria & Sarah S.Nicholls, The Emperor's New Drugs: An Analysis of Antidepressant Medication DataSubmitted to the U.S. Food and Drug Administration, PREVENTION & TREATMENT, July 2002,at 3 [hereinafter Kirsch et al., Analysis ofAntidepressant]; Irving Kirsch, Antidepressants: The

[Vol. 94:773


legal standard only requires there be "substantial evidence" of the "intended effect,"with the intended effect defined by the company.33 Importantly, the full results ofclinical studies are generally not available to the public including researchers.34 Eventhough companies may publish scientific articles relating to approved drugs, theygenerally selectively publish positive results,35 and even then, the articles are likelyfar less detailed than the clinical data available to the FDA.36 Accordingly, it is not

Emperor's New Drugs?, HUFFPOST (Nov. 17, 2011), https://www.huffpost.com/entry/antidepressants-the-emper b_442205 [https://perma.cc/P5RH-9CTZ]. Moreover, an FDAmemo concerning Celexa noted that two controlled efficacy trials showed significantdifferences between the drug and placebo whereas three others "[flailed to provide resultsconfirming the positive findings," but the FDA nonetheless concluded that "there is clearevidence from more than one adequate and well controlled clinical investigation' that the drughad an antidepressant effect. Kirsch et al., Analysis ofAntidepressant, supra. About a third ofdrugs are approved based on a single pivotal trial. Nicholas S. Downing, Jenerius A.Aminawung, Nilay D. Shah, Harlan M. Krmholz & Joseph S. Ross, Clinical Trial EvidenceSupporting FDA Approval of Novel Therapeutic Agents 2005-2012, 311 J. AM. MED. ASS'N368, 372 (2014).

33. 21 U.S.C. § 355(d)(5)(iv) (2012); see also Jonathan J. Darrow, PharmaceuticalEfficacy: The Illusory Legal Standard, 70 WASH. & LEE L. REV. 2073 (2013) [hereinafterDarrow, Pharmaceutical Efficacy] (criticizing FDA approval standard for lacking any specificlevel of efficacy and instead simply mandating precise calculation of efficacy). Traditionally,the FDA required at least two clinical trials to suggest a drug is superior to a placebo orconventional treatment. FOOD & DRUG ADMIN., GUIDANCE FOR INDUSTRY: PROVIDING

CLINICAL EVIDENCE OF EFFECTIVENESS FOR HUMAN DRUGS AND BIOLOGICAL PRODUCTS 3(1998). However, there could be other contrary studies. This was definitely true for a numberof antidepressants. See supra note 32. In addition, as a result of congressional amendments,the FDA can now approve a drug based on positive results from a single trial. 21 U.S.C. §355(d).

34. The FDA has traditionally considered clinical studies to be a trade secret and fallwithin an exemption to FOIA requests and courts have concurred. Judicial Watch, Inc. v. Food& Drug Admin., 449 F.3d 141, 149 (D.C. Cir. 2006) (finding that FDA was entitled towithhold some data sought by FOIA request regarding approved drug RU-486 for medicalabortion based on the exemption which covers some information contained in NDAs); PublicInformation, 42 Fed. Reg. 3093, 3094 (Jan. 14, 1977) (noting FDA has treated clinical trialdata as trade secret since 1938); 21 C.F.R. § 20.85 (1994); see also 21 U.S.C. § 355(1) (creatingan exemption to FOIA requests in the event of likelihood of substantial competitive harm).However, the FDA position on trade secrecy of clinical data based on 21 U.S.C. § 331() hasbeen criticized since clinical data are arguably not a method or process. E.g., Christine D.Galbraith, Dying to Know: A Demand for Genuine Public Access to Clinical Trial ResultsData, 78 MISS. L.J. 705, 752-42 (2009). In addition, recent laws permit proactive disclosureby the FDA E.g., 21 U.S.C. § 355(r) (instructing FDA to have website to provide better accessto drug safety information for patients and providers and mandating FDA post safety alerts,warning letters, as well as "other material determined appropriate" by the FDA); AmyKapczynski & Jeanie Kim, Clinical Trial Transparency: The FDA Should and Can Do More,45 J.L. MED. & ETHICS 33, 35 (2017) (noting that FDA should be routinely releasing clinicaldata, as well as summary results as "other material"). Moreover, even though clinical studieshave recently been obtained through FOIA, that was notably a highly time-consuming process.E.g., Kapczynski & Kim, supra.

35. See infra notes 46, 59-60 and accompanying text (discussing selective publication).36. E.g., Kapcyzinski & Kim, supra note 34, at 34.

2019]

INDIANA LAW JOURNAL

unusual for problems with drugs to be discovered years later by independentresearchers who must do their own tests. These researchers may find out that the drugdoes not work as advertised and sometimes even works contrary to how it isadvertised.37 These issues have prompted calls for full disclosure of clinical trialresults.38 Although the FDA does not currently require complete disclosure, it doesrequire companies that seek FDA approval of drugs to publicly register details of

37. For example, for years, hormone replacement drug therapy was promoted to not onlytreat menopause, but also to prevent heart disease based on industry-financed data untilindependent researchers at the NIH discovered that the drugs actually increased the risk ofheart disease. Nancy Krieger et al., Hormone Replacement Therapy, Cancer, Controversies,and Women's Health: Historical, Epidemiological, Biological, Clinical, and AdvocacyPerspectives, 59 J. EPIDEMIOLOGY & COMMMUNITY HEALTH 740, 740 (2005); see also Lisa A.Ladewski et al., Dissemination of Information on Potentially Fatal Adverse Drug Reactionsfor Cancer Drugs from 2000 to 2002: First Resultsfrom the Research on Adverse Drug Eventsand Reports Project, 21 J. CLINICAL ONCOLOGY 3859 (2003) (discussing serious, includingfatal, adverse drug reactions that are only discovered years after FDA approval); Heidi D.Nelson, Miranda Walker, Bernadette Zakher & Jennifer Mitchell, Menopausal HormoneTherapy for the Primary Prevention of Chronic Conditions: A Systematic Review to Updatethe U.S. Preventative Services Task Force Recommendations, 157 ANNALS INTERNAL MED.

104, 109-10 (2012) (systematic review of studies that confirms increased risk of stroke).38. E.g., Peter Doshi, Tom Jefferson & Chris Del Mar, The Imperative to Share Clinical

Study Reports: RecommendationsJrom the Tamiflu Experience, PLOS MED., Apr. 2012, at 1-2 (noting that clinical study reports have the same information as journal papers, but moredetail); Ben Goldacre, What the Tamiflu Saga Tells Us About Drug Trials and Big Pharma,GUARDIAN (Apr. 10, 2014, 2:00 PM), https://www.theguardian.com/business/2014/apr/10/tamiflu-saga-drug-trials-big-pharma [https://perma.cc/4DMV-CG88]; Katie Thomas,Breaking the Seal on Drug Research, N.Y. TIMES (June 29, 2013), https://www.nytimes.com/2013/06/30/business/breaking-the-seal-on-drug-research.html [https://perma.cc/TT5F-8EBH]. These suggestions were prompted by eventual public disclosure that off-labelpromotion of Tamiflu to reduce hospital complications was found to be unjustified based oncomplete data, but only two often studies were published and none of published data disclosednegative side effects. Yogendra Gupta, Meenaskshi Meenu & Prafull Mohan, The TamifluFiasco and Lessons Learnt, 47 INDIAN J. PHARMACOLOGY 11 (2015); Shannon Brownlee &Jeanne Lenzer, The Truth About Tamiflu, ATLANTIC (2009), https://www.theatlantic.com/magazine/archive/2009/12/the-truth-about-tamiiflu/307801 [https://perma.cc/8GL6-ZGP3];see also Jeanne Lenzer, Conflicting (Conflicted?) Info, Tamiflu and Unquestioning NewsReporting, HEALTHNEwSREVIEW (Feb. 10, 2015), https://www.healthnewsreview.org/2015/02/conflicting-study-reports-tamiflu-and-unquestioning-news-reporting [https://perma.cc/DRS7-XYWB] (noting that the Centers for Disease Control and Prevention (CDC)promotion of off-label use of Tamiflu was based on a supposedly independent article, but threequarters of the authors received funding).

[Vol. 94:773


clinical trials when initiated, as well as provide summary results.3 9 However,companies thus far have not fully complied with even these limited disclosures.4"

B. Marketing of Prescription Drugs

When a prescription drug is initially approved by the FDA, the only source ofinformation regarding the drug is provided by the self-interested drug company.

39. 42 U.S.C. § 282(j)(2)(C)((ii)) (2012) (requiring new clinical trials to submit the trialinformation to a registry no later than 21 days after first patient is enrolled); 42 C.F.R. §11.24(a) (2016) (requiring submission of clinical trial registration); 81 Fed. Reg. 64,983 (Sept.21, 2016) (noting summary data must be submitted in tabular form with key informationincluding primary and secondary outcomes); see also Darrow, Pharmaceutical Efficacy, supranote 33, at 2101 (noting a 2007 change to FDA laws that required all clinical trials to beincluded as part of Food and Drug Administrative Amendments Act, instead of only seriousor life-threatening conditions under 1997 legislation). In addition, the FDA recentlyannounced a pilot program to reveal summary data concerning pivotal trials, although this ison a voluntary basis and still does not reveal all underlying clinical data. Press Release, FDA,FDA Commissioner Scott Gottlieb, M.D., on New Steps FDA Is Taking to EnhanceTransparency of Clinical Trial Information to Support Innovation and Scientific InquiryRelated to New Drugs (Jan. 16, 2018), https://www.fda.gov/newsevents/newsroom/pressannouncements/ucm592566.htm [https://perma.cc/6FYV-VP46]. In contrast, Europe ismoving towards proactive publication of clinical trial data, even without a specific request.EUROPEAN MEDS. AGENCY, EUROPEAN MEDICINES AGENCY POLICY ON ACCESS TO DOCUMENTS

(RELATED TO MEDICINAL PRODUCTS FOR HUMAN AND VETERINARY USE) (2010); EUROPEAN

MEDS. AGENCY, EUROPEAN MEDICINES AGENCY POLICY ON PUBLICATION OF CLINICAL DATA

FOR MEDICINAL PRODUCTS FOR HUMAN USE (2014) (permitting online publication of allclinical study reports on a proactive basis since most information is not confidential); see alsoCommission Regulation 726/2004, art. 80, 2004 (EC) (providing basis for Policy 0070);Commission Regulation 536/2014, art. 81, 2014 (EC) (requiring all clinical trials conductedin the European Union to be made available to the public through a publicly available databasealthough with the possibility for some commercially confidential information to be withheld);see also Elisa Stefanini, Publication of Clinical Trials Data: A New Approach to Transparencyin the European Legislative Framework, POINT CARE (2017) (explaining the change inregulations). The EU Regulation is expected to come into effect in 2019. EuropeanCommission, Clinical Trials Regulation EU No. 53612014,https ://ec.europa.eu/health/human-use/clinical-trials/regulation en [https ://perma.cc/P76T-HECC].

40. Current reporting is just under sixty percent. Who's Sharing Their Clinical TrialResults?, FDAAA TRIALSTRACKER, http://fdaaa.trialstracker.net/?status / 5B / 5D=overdue&status%5B%5D=overdue-cancelled&status%5B%5D=reported-late [https://perma.cc/G6RD-NSUY]. Although this is not ideal, it is an improvement. Andrew P. Prayle, MatthewN. Hurley & Alan R. Smyth, Compliance with Mandatory Reporting of Clinical Trial Resultson ClincialTrials.gov: Cross Sectional Study, 344 BRIT. MED. J. 3 (2012) (finding seventy-eight percent of trials from 2009 were not reported). In addition, although the FDA couldenhance compliance by imposing statutorily permitted penalties of up to $10,000 a day, it hasyet to do so. FDAAA TrialsTracker, supra (interactive and constantly updated website notinguncollected fines of over $300 billion to date); Ben Goldacre, Health Care's Trick Coin, N.Y.TIMES (Feb. 1, 2013), https://www.nytimes.com/20 13/02/02/opinion/health-cares-trick-coin.html [https://perma.cc/6EVD-367Y] (noting no fines levied for violation); see also 21U.S.C. § 333(f)(3)(B) (2012) (authorizing penalties of up to $10,000 a day).

2019]

INDIANA LAW JOURNAL

Unlike other consumer goods where there are competing manufacturers in the sameclass that compare their products to each other, or evaluations by unbiased thirdparties, that does not happen with these drugs.41 This is partially because FDAapproval is generally based on establishing safety and efficacy, but not on superioritycompared to other drugs.42 So, companies often provide data only comparing theirdrug to a placebo, known to have no effect, rather than existing treatments.43 Then,each company that makes and sells one drug in the same class of drugs, such as thosethat treat heartburn, will market its drug as effective, but generally without any clearindication to doctors or consumers of which drug is better. In addition, in some caseswhere a company markets its drug as better than a competitor's, the marketing mayfail to emphasize key limitations of the comparison, such as the fact that thesupporting study compared a weaker dose of the competitor's drug. 44

In the unique marketplace of prescription drugs, there is initially no unbiasedinformation and advertising looms large in spreading information that may later bedetermined to be false. Companies can easily present misleading information by onlyadvertising positive information about drugs. In some cases, it is subsequentlyrevealed that companies intentionally concealed known negative information orselectively only presented positive results in marketing45 and academic

41. For example, although consumers rely on sources like the magazine ConsumerReports, as well as other third-party reviews, nothing similar exists in the more complex areaof prescription drugs since reviews are more complex than evaluating goods such as appliancesor cars. There are some independent newsletters that provide doctors with information aboutnew drugs. E.g., HEALTH ACTION INT'L, FACT OR FICTION?: WHAT HEALTHCARE

PROFESSIONALS NEED TO KNOW ABOUT PHARMACEUTICAL MARKETING IN THE EUROPEANUNION 52-53 (noting independent sources such as the monthly Prescriber's Letter concerningnew treatments as well as "Worst Pills, Best Pills" that provides information on prescriptionand over-the-counter drugs). However, doctors do not seem to refer to these in studiesregarding sources of information.

42. See 21 U.S.C. § 355(d) (2012); 21 C.F.R. § 202.1 (2017).43. Eg., MARCLA ANGELL, THE TRUTH ABOUT DRUG COMPANIES 112 (2004); see also

CENTER FOR INFORMATION AND STUDY ON CLINICAL RESEARCH PARTICIPATION, WHAT IS A

PLACEBO CONTROLLED CLINICAL TRIAL? (explaining placebo controlled clinical trials); GaryT. Chiodo, Susan W. Tolle & Leslie Bevan, Placebo-Controlled Trials: Good Science orMedical Neglect?, 172 W.J. MED. 271, 271 (2000) (noting that the gold standard for clinicaltrials has traditionally been randomized trials comparing experimental treatment withplacebo). However, placebos are not typically used in studying treatments regarding seriousand life-threatening conditions including cancer. Clinical Research Versus MedicalTreatment, FOOD & DRUG ADMIN., (Mar. 22, 2018)https ://www.fda.gov/forpatients/clinicaltrials/clinicalvsmedical/default.htm [https ://perma.cc/7MS7-2H8U].

44. ANGELL, supra note 43, at 78-79. Although this is not helpful to consumers,corporations have incentives to not only promote their bottom line, but to avoid clinical studiesthat end up helping competitors. E.g., Winslow, supra note 11 (discussing Bristol MyersSquibb's failed attempt to establish that its cholesterol-lowering drug compared well toPfizer's Lipitor in a rare head to head comparison of competing drugs).

45. Sometimes companies instruct sales representatives to not disclose relevantinformation. E.g., Shannon Hall & Jeanne Lenzer, The Problem with Medicine: We Don'tKnow if Most of It Works, DISCOVER (Feb. 11, 2011), http://discovermagazine.com/2010/nov/11-the-problem-with-medicine-dont-know-if-most-works [https://perma

[Vol. 94:773


publications;46 there are even reports that companies sometimes threatenindependent doctors or scientists who raise issues."' Although companies can be

.cc/A8JG-VBDE] (noting that Pfizer representatives were instructed not to provide relevantstudy showing that inexpensive generic drugs were equally effective as brand name drugs forreducing heart attacks and actually safer). At other times, sales representatives are instructedto provide messages that are not firmly supported by data. For example, Purdue trained itssales representatives to tell doctors the risk of addiction from OxyContin was less than onepercent. This was based on a study involving burn victims treated with opioids that had nobearing on the long term use of opioids to treat chronic pain, and in fact, contradicts manystudies showing high incidence of abuse for chronic pain. Art Van Zee, The Promotion andMarketing of OxyContin: Commercial Triumph, Public Health Tragedy, 99 AM. J. PUB.HEALTH 221, 223 (2009).

46. A famous example involves the anti-inflanunatory drug sold as Vioxx, that was at onepoint the best-selling drug with over $2 billion a year in sales-before it was removed fromthe market after initially suppressed studies revealed that it was associated with an increase inheart attacks and strokes. HOLLY PRESLEY, INSTITUTIONS IN CRISIS: VIOXX AND THE MERCK

TEAM EFFORT (2009); see also ELLIOTT, supra note 20, at 103 (noting that although companydisclosed study showing that Vioxx led to a 500% increase of risk of heart attacks comparedto naproxen, Merck stated that naproxen protected the heart); Rita Rubin, How Did VioxxDebacle Happen?, USA TODAY (Oct. 12, 2004, 12:00 AM), https://usatoday30.usatoday.com/news/health/2004-10-12-vioxx-cover x.htm [https://perma.cc/SD2M-7789].In some cases, companies have been found to fail to adequately notify the public of knownrisks. E.g., Press Release, U.S. Senate Comm. on Fin., Grassley, Baucus, Release CommitteeReport on Avandia, (Feb. 20, 2010), https://www.finance.senate.gov/release/grassley-baucus-release-conuittee-report-on-avandia [https ://perma.cc/T5 CQ-WA9A] (discussing Senate Report concluding that GSK failed to warn patients regardingcardiovascular risks from its drug Avandia and that the company instead improperlyintimidated independent doctors to not discuss problems). Along somewhat similar lines,companies sometimes financially support entities that appear independent to the public, butessentially advocate the corporate position; these could be seemingly independentassociations, as well as patient advocacy groups. E.g., Celine Grounder, Who Is Responsiblefor the Pain-Pill Epidemic?, NEW YORKER (Nov. 8, 2013), https://www.newyorker.com/business/currency/who-is-responsible-for-the-pain-pill-epidemic [https://perma.cc/LZK9-TN9A] (noting the role of the Joint Commission, which controls accreditation of healthfacilities and was funded by companies to penalize physicians for undertreatment of pain thathelped to encourage more use of opioids); Charles Ornstein & Tracy Weber, American PainFoundation Shuts Down as Senators Launch Investigation of Prescription Narcotics,PROPUBLICA (May 8, 2012, 8:57 PM), https://www.propublica.org/article/senate-panel-investigates-drug-company-ties-to-pain-groups [https ://perma.cc/R4BK-4HV5] (notingfinancial ties between the now defunct American Pain Foundation and the industry); CharlesOrnstein & Tracy Weber, The Champion of Painkillers, PROPUBLICA (Dec. 23, 2011, 9:15AM), https://www.propublica.org/article/the-champion-of-painkillers [https://perma.cc/EA9V-Q5UT] (noting American Pain Society's role in encouraging aggressive treatment ofpain, including use of narcotics).

47. E.g., ELLIOTT, supra note 20, at 103-04 (noting Merck attempts to neutralize anddiscredit those that raised concerns regarding Vioxx and heart disease included threateningwithdrawal of funding to researcher's university); Gardiner Harris, Research Ties DiabetesDrug to Heart Woes, N.Y. TIMES (Feb. 19, 2010), https://www.nytimes.com/2010/02/20/healtl/policy/20avandia.html [https://perma.cc/4PEU-SVF6] (noting thatGlaxoSmithKline intimidated independent doctors and investigators with potential legal

2019]

INDIANA LAW JOURNAL

sanctioned,4" the reality is that before problems are discovered, public health may beat risk for a number of years.4 9 Moreover, public health may suffer even afterindependent scientists debunk pharmaceutical claims; a pivotal example is the opioidepidemic that resulted from aggressive industry marketing that includedaffirmatively false information, such as the assertion that opioids were not addictive,which resulted in a settlement of criminal charges in 2007, as well as more recentlitigation for continuing misrpresentation0

Companies seem well aware of the relevance and impact of their advertising sincethey spend substantial resources advertising not only to consumers in the United

action for suggesting heart risks with its drug Avandia even though the company was awareof risks, yet continued to market the drug aggressively).

48. For example, Paxil was sued for fraud for concealing negative information aboutPaxil, that included four negative clinical trials and only one positive one. Complaint, UnitedStates v. GlaxoSmithKline (D. Mass. Oct. 26, 2011) (No. 11-10398-RWZ) (suit for falseclaims act).

49. For example, Paxil was sued for fraud for concealing negative information aboutPaxil, that included four negative clinical trials and only one positive one. Complaint, UnitedStates v. GlaxoSmithKline (D. Mass. Oct. 26, 2011) (No. 11-10398-RWZ) (suit for falseclaims act). Gupta, supra note 38, at 11, 13 (noting that serious adverse events were reportedafter approval of Tamiflu that were not in the selectively published articles). Indeed, onescholar has suggested that newly approved FDA drugs are de facto tested on the general publicsince it is impossible for "clinical trials to detect rare adverse events." Jonathan J. Darrow,Crow dsourcing Clinical Trials, 98 MNN. L. REv. 805, 805 (2014).

50. United States v. Purdue Frederick Co., 495 F. Supp. 2d 569 (W.D. Va. 2007);Complaint, Oregon v. Purdue (Or. Cir. Ct., Sept. 18, 2018), https://www.doj.state.or.us/wp-content/uploads/2018/09/State v. Purdue Complaint.pdf [https://perma.cc/ZK3H-WGND];Barry Meier, In Guilty Plea, OxyContin Maker to Pay $600 Million, N.Y. TiMEs (May 10,2007), https://www.nytimes.com/2007/05/10/business/1 ldrug-web.htnml [https://perma.cc/D7DH-585S]; Barry Meier, Origins ofan Epidemic: Purdue Pharma Knew its Opioids WereWidely Used, N.Y. TMEs (May 29, 2018), https://www.nytimes.com/2018/05/29/health/purdue-opioids-oxycontin.htmil [https://perma.cc/RK2V-GB4M]; Joanna Walters,Oxycontin Maker Expected 'A Blizzard of Prescriptions' Following Drug's Launch,GUARDIAN (Jan 16, 2019, 1:00 PM), https://www.theguardian.com/us-news/2019/jan/15/oxycontin-purdue-pharma-massachusetts-opioid-crisis [https://perma.cc/94F7-238U]; KatieZezima & Lenny Berstein, Hammer on the Abusers ': Mass. Attorney General Alleges PurduePharma Tried to Shift Blame for Opiod Addicition, WASH. POST (Jan 15, 2019), https://www.washingtonpo st.com/national/hanuner-on-the -abusers -mas s-attorney-general-alleges-purdue-pharma-tried-to-shift-blame-for-opioid-addiction/2019/01/15/4af25c4c- 190c-1 1e9-88fe-f9f77a3bcb6cstory.html?utm term=.40c054446211 [https://perma.cc/3RNP-ZM92].

[Vol. 94:773


States,51 but also especially to doctors.5 2 Most money is spent on "detailing," whichrefers to work done by individual sales representatives who visit doctors topersonally inform them of new drugs with presentations and brochures, as well asprovide samples.53 Unsurprisingly, this information has been found to be incompleteand misleading.54 Although doctors generally recognize that detailing is likelybiased, they may still rely on it because it is convenient.55 Along somewhat similarlines, companies have developed relationships with doctors who are perceived as keyopinion leaders to influence peers and even provide specific presentation slides for

51. The United States is one of two countries permitting companies to advertise drugsdirectly to consumers (the other being New Zealand); in both cases, such advertising is ofrelatively recent vintage. Sandra Coney, Direct-to-Consumer Advertising of PrescriptionPharmaceuticals. A Consumer Perspective from New Zealand, 21 J. PUB. POL'Y &MARKETING 213, 213-14 (2002) (explaining that New Zealand laws were not explicitlywritten to cover advertising, but the industry exploited them in the late 1980s and despitesubsequent proposal to ban such advertising, it has remained). Spending on such ads has sincegrown substantially. Avinandan Mukherjee, Yam Limbu & Isaac Wanasika, A Review ofResearch on Direct-to-Consumer Advertising of Prescription Drugs. Directions for FutureResearch, 7 INT'L J. PHARMACEUTICAL & HEALTHCARE MARKETING 226, 226-27 (2013)(noting a 330% increase between 1996 and 2005).

52. See Lisa M. Schwartz & Steven Woloshin, Medical Marketing in the United States,1997-2016, 32 J. AM. MED. ASS'N 80, 82, 87 (2019) (finding that marketing to medicalprofessionals accounted for the highest proportion of spending during the time studied);Persuading the Prescribers. Pharmaceutical Industry Marketing and Its Influence onPhysicians and Patients, PEW (Nov. 11, 2013), https://www.pewtrusts.org/en/research-and-analysis/fact-sheets/2013/11/11/persuading-the -prescribers -pharmaceutical-industry-marketing-and-its-influence-on-physicians-and-patients [https://perma.cc/75CU-EJBD] (notingthat the industry spent $27 billion in 2012, including more than $24 billion marketing todoctors alone); Nicole Van Groningen, Big Pharma Gives Your Doctor Gifts. Then YourDoctor Gives You Big Pharma's Drugs, WASH. POST (June 13, 2017), https://www.washingtonpost. com/opinions/big-pharma-gives-your-doctor-gifts-then-your-doctor-gives-you-big-pharmas-drugs/2017/06/13/5bc0b550-5045-1 1e7-b064-828ba60fbb98_story.html?noredirect on&utm term=.b80b lf9ee8fd [https://perma.cc/XT7V-URWX] (noting that theindustry spent $24 billion marketing to health care professionals in 2012, which is eight timesthe amount spent advertising to consumers). It is estimated that the industry spends $8000-$13,000 per year on an individual doctor. See Wazana, supra note 20, at 373.

53. Marc-Andrd Gagnon & Joel Lexchin, The Cost of Pushing Pills. A New Estimate ofPharmaceutical Promotion Expenditures in the United States, PLoS MED., Jan. 2008, at 29-31 (finding that detailing and samples constituted the majority of marketing expenditures); seealso Schwartz & Woloshin, supra note 52, at 84, 86 (finding the cost of detailing to doctorsfar exceeded medical journal advertising and that companies spent even more on the cost ofsamples).

54. E.g., Smita Mali, Sujata Dudhgaonkar & N. Bachewar, Evaluation of Rationality ofPromotional Drug Literature Using World Health Organization Guidelines, 42 INDIAN J.PHARMACOLOGY 267 (2010). John Oliver has exposed this problem to the public at large.Marketing to Doctors. Last Week Tonight with John Oliver (HBO), YouTUBE (Feb. 8, 2015),https://www.youtube.com/watch?v=YQZ2UeOTO31 [https://perma.cc/85EE-RJ88].

55. See HEALTH ACTION INT'L, supra note 41, at 13 (noting that health professionals areshort on time and thus may be vulnerable to marketing); see also infra note 168 andaccompanying text (noting that some doctors' belief in the value of marketing is so strong thatthey joined a lawsuit against the FDA to permit more marketing).

2019]

INDIANA LAW JOURNAL

such opinion leaders.56 In addition, companies finance scientific studies to bepublished in peer reviewed journals to support their drugs.5 7 Although peer reviewjournals are often presumed to be unbiased, especially by courts,58 in recent years,studies have found problems with such articles since there is a general publicationbias towards positive results, with some results incapable of reproduction.5 9

Moreover, studies published by self-interested companies are four times more likelyto have favorable outcomes.6" This is not surprising since the industry has no interestin publishing unfavorable data; so, even though the FDA has all the data, the publicknowledge is incomplete. There are some notable instances of public harm duringthe period after drug approval when independent scientists conduct their own studieswithout access to original data.61 In addition, the incomplete public knowledge from

56. E.g., Sergio Sismondo, Key Opinion Leaders and the Corruption of MedicalKnowledge: What the Sunshine Act Will and Won't CastLight on, 41 J.L. MED. & ETHICS 635(2013). Of course, after the Sunshine Act, payments to physicians need to be disclosed, whichmight limit payments. However, this nonetheless is an example of how companies are investedin utilizing all possible avenues of influencing doctors, including methods that other doctorsmay not perceive as marketing.

57. E.g., ELLIOTT, supra note 20, at 28-39 (discussing the ghost writing of articles); JoelLexchin, Those Who Have the Gold Make the Evidence: How the Pharmaceutical IndustryBiases the Outcomes of Clinical Trials of Medications, 18 Sci. & ENGINEERING ETHICS 247(2011). Journals have their own self-interest to consider when publishing articles from theindustry that sponsors lucrative ads. See ELLIOTT, supra note 20, at 39 (noting that in 2004medical journals generated twice as much from advertising revenue as from subscriptions);Kamran Abbasi & Richard Smith, No More Free Lunches, 326 BRIT. MED. J. 1155 (2003)("[T]he stark reality is that without pharmaceutical sponsorship many journals would notsurvive."). Moreover, companies that publish articles will also pay hefty fees for reprints ofthe articles to then distribute to doctors. ELLIOTT, supra note 20, at 41.

58. E.g., Wash. Legal Found. v. Friedman, 13 F. Supp. 2d 51, 67 (D.D.C. 1998) (notingthat the "FDA exaggerates its overall place in the universe" in suggesting that FDA reviewwas necessary to ensure accuracy of data in peer review publications that companies wantedto promote).

59. E.g., John P. A. Joannidis, Why Most Published Research Findings Are False, PLOSMED., Aug. 2015, at 0696, 0697; SILAS BOYE NISSEN, TALI MAGIDSON, KEVIN GROSS & CARL

T. BERGSTROM, ELIFE, PUBLICATION BIAS AND THE CANONIZATION OF FALSE FACTS 1, 13(2016).

60. Joel Lexchin, Lisa A. Bero, Benjamin Djulbegovic & Otavio Clark, PharmaceuticalIndustry Sponsorship and Research Outcome and Quality: Systematic Review, 326 BRIT. MED.

J. 1167 (2003); Richard Smith, Medical Journals Are an Extension of the Marketing Arm ofPharmaceutical Companies, PLoS MED., May 2005, at 364; see also Donald W. Light, JoelLexchin & Jonathan J. Darrow, Institutional Corruption of Pharmaceuticals and the Myth ofSafe and Effective Drugs, 14 J.L. MED. & ETHICS 590, 595 (2013) (noting that positive resultscan be published twice). This can be done "by asking the 'right' questions," such as comparingthe drug with a treatment known to be inferior, or low doses of competitor drugs, as well aspublishing the same results multiple times. Smith, supra, at 365; see also Christopher T.Robertson, The Money Blind. How to Stop Industry Bias in Biomedical Science, WithoutViolating the First Amendment, 37 AM. J.L. & MED. 358, 370-71 (2013).

61. Vioxx is an excellent example of this problem since a published article failed todisclose all data and misrepresented heart risks. Harlan Krumholz et al., What Have We Learntfrom Vioxx?, 334 BRIT. MED. J. 120, 121 (2007).

[Vol. 94:773


published literature is further complicated by the fact that some seeminglyindependent articles may actually be ghost written by the industry.6 2

Studies repeatedly show that pharmaceutical promotion efforts do influencewhich drugs doctors prescribe. Some studies indicate that doctors who rely more onmarketing information tend to prescribe more.63 Meetings with pharmaceuticalrepresentatives impact prescriptions, as well as doctor requests for drugs to be addedto formularies, even when not superior to existing drugs.64 Pharmaceutical marketingcan be highly effective at driving demand for drugs, even when those drugs are notsuperior to existing drugs.65 The effectiveness of marketing to persuade doctors is

62. ELLIOTT, supra note 20, at 28-39 (discussing ghost-writing articles).63. E.g., Roberto Cardarelli, John C. Licciardone & Lockwood G. Taylor, A Cross-

Sectional Evidence-Based Review of Pharmaceutical Promotional Marketing Brochures andTheir Underlying Studies: Is What They Tell Us Important and True?, BMC FAM. PRAc., Mar.2006, at 1, 2; T. Shawn Caudill, Nicole Lurie & Eugene C. Rich., The Influence ofPharmaceutical IndustryAdvertising on Physician Prescribing, 22 J. DRUG ISSUES 331 (2012).

64. Wazana supra note 20, at 375 (noting multiple studies showing that meeting withpharmaceutical representatives was associated with increased prescriptions and formularyrequests and that these drugs generally presented no therapeutic advantage); see also Wash.Legal Found. v. Friedman, 13 F. Supp. 2d 51, 63-64 (D.D.C. 1998) (noting substantialevidence that meeting with representatives increased prescription sales, such thatmanufacturers would want to disseminate information even though the court ultimately foundit impermissible for the FDA to limit information from representatives). The impact ofdetailing is also underscored by the fact that when detailing is limited, prescriptions fall. IanLarkin, Desmond Ang, Jerry Avom & Aaron S. Kesselheim, Restrictions on PharmaceuticalDetailing Reduced Off-Label Prescribing of Antidepressants and Antipsychotics in Children,33 HEALTH AFF. 1014, 1020-21 (2014) [hereinafter Larkin et al., Restrictions] (finding thatafter strict detailing policies, there was a reduction in market share of more than one-third ofdrugs detailed and approved for use in children and even for uses unapproved in children, eventhough these should theoretically never have been promoted); see also Ian Larkin et al.,Association Between Academic Medical Center Pharmaceutical Detailing Policies andPhysician Prescribing, 317 J. AM. MED. ASS'N 1785 (2017) [hereinafter Larkin et al.,Physician Prescribing] (finding association between limiting representatives and change inprescribing).

65. E.g., Cardarelli et al., supra note 63, at 2 (noting that marketing of expensive newdrugs using calcium channel blockers resulted in the new drugs commanding a greater marketshare than older and equally effective drugs); Hall & Lenzer, supra note 45 (quoting Dr.Hoffman, a professor of emergency medicine at the University of Southern California, whosays that repeatedly "[s]ome expensive new drug becomes a blockbuster best seller followingextensive marketing, even though the best one might be able to say about it is that it seemsstatistically 'non-inferior' to an older, cheaper drug"). Moreover, this can lead to public healthproblems, such as with Vioxx, which was not only not better than prior drugs, but actually lesssafe, as well as OxyContin, which provided no advantage over existing opioids, but aggressivepromotion has promoted an opioid addiction crisis. Van Zee, supra note 45, at 221-24 (notingstudies that indicate the drug's only improvement is reduced dosing, but that marketing offalse information, together with generous bonuses to representatives for such marketing,resulted in a nearly tenfold increase of prescriptions for non-cancer pain relief, as well asaddiction). Indeed, OxyContin was the most prescribed brand name opioid in 2001 formoderate to severe pain, even though it had not been shown to be superior to other opioids.Id. at 225.

2019]

INDIANA LAW JOURNAL

not surprising since companies have relied heavily on methods known in marketingliterature to be persuasive.66 Moreover, companies also indirectly influence doctorsby marketing directly to consumers. Studies have shown that some doctors feelpressure to prescribe a specific drug requested by a consumer, even when a doctormay not be personally persuaded that it is a superior treatment.6 7 Some policiesintended to limit pharmaceutical company interaction with physicians to decreaseconflicts of interest have resulted in changes to some prescription behavior.68

C. FDA Regulation of Off-Label Uses

An important subset of pharmaceutical marketing involves the extent to whichcompanies are permitted to regulate so-called "off-label" uses of drugs that the FDAhas approved. This Section briefly explains what off-label use is, and then explainsthe current FDA position on regulation of the marketing of such uses.

The first question is what constitutes an "off-label" use. As previously discussed,the FDA approves a specific drug not for any purpose, but only for intendedpurpose(s) that a company requests. So, when a company requests FDA approval ofa drug, it focuses its clinical tests on the intended purpose(s) it wishes to establish assafe and effective, rather than all possible purposes. Notably, although FDA approvalis based on particular intended use(s), once approved, a drug may be prescribed by a

66. E. E. Roughead, K. J. Harvey & A. L. Gilbert, Commercial Detailing TechniquesUsed by Pharmaceutical Representatives to Influence Prescribing, 28 AUSTL. & N.Z. J. MED.306, 307 (1998) (techniques include reciprocity for gifts, friendship, social validation, appealto authority, and scarcity); Allen F. Shaughnessy, David C. Slawson & Joshua H. Bennett,Separating the Wheat from the Chaff Identifying Fallacies in Pharmaceutical Promotion, 9J. GEN. INTERNAL MED. 563 (1994) (noting that although information on its own almost neverpersuades, all advertising, including advertisement of drugs, is successful by appealing toemotions and vulnerabilities).

67. E.g., Barbara Mintzes, Morris L. Barer, Richard L. Kravitz, Ken Bassett, JoelLexchin, Arminde Kazanjian, Robert G. Evans, Richard Pan & Stephen A. Marion, How DoesDirect-to-Consumer Advertising (DTCA) Affect Prescribing? A Survey in Primary CareEnvironments' with and Without Legal DTCA, 169 CANADIAN MED. Ass'N J. 405, 408, 411(2003) (finding that advertising leads to patient requests for drugs and that doctors in roughlyhalf of cases prescribed a drug that was "unlikely" or only "possibly" relevant); Joel S.Weissman, David Blumenthal, Alvin J. Silk, Michael Newman, Kinga Zapert, Robert Leitman& Sandra Feibelmann, Physicians Report on Patient Encounters Involving Direct-to-Consumer Advertising, HEALTH AFF. (WEB EXCLUSIvE) 219, 225 (2004) (finding that thirty-nine percent of doctors prescribed advertised drugs, with less than half believing the prescribeddrug was the most effective drug for the patient). Some have suggested that doctors yield topatient requests to avoid disappointment, as well as switching physicians. E.g., Robert A. Bell,Michael S. Wilkes & Richard L. Kravitz, Advertisement-InducedPrescription Drug Requests:Patients 'Anticipated Reactions to a Physician Who Refuses, 48 J. FAM. PRAC. 446 (1999).

68. Larkin et al., Physician Prescribing, supra note 64, at 1786, 1793 (2017) (providingan evaluation of U.S. academic medical centers in five states with largest number ofacademically affiliated physicians found eight of eleven institutions with policies regulatinggifts and representative access to have statistically significant results, as well as statisticallysignificant changes in market share for six of eight drug classes).

[Vol. 94:773


doctor for any purpose, including uses not approved by the FDA.69 In other words,when a doctor prescribes an approved drug for a different purpose or use, it isconsidered "off-label."

At first glance, it may seem inefficient for the FDA to only evaluate a drug basedon one, or even a few limited, intended purposes. However, the FDA regulatorysystem involves a delicate balance. The FDA wants to protect public health and alsowants the regulatory system to incentivize companies to develop and submit data tosupport specific uses. So, for example, a company that obtains approval for one initialuse of a drug can later request approval for additional uses-if it provides dataestablishing these new uses as safe and effective." However, since FDA jurisdictioncovers drugs, but not the practice of medicine, the FDA does not bar doctors fromprescribing drugs for off-label use.71

The FDA did not initially set out to regulate off-label uses, or marketing of suchuses. Rather, it only considered regulating off-label use after a specific off-label usetouted in a journal article led to serious concern that ultimately promptedcongressional pressure for the FDA to rigorously regulate off-label use. 2 The FDAinitially proposed direct regulation of off-label use, but after opposition from doctors,the FDA asserted it would not limit medical treatment and nixed a proposal to barshipments of drugs for off-label use.73 Since then, the FDA has developed its currentposition that focuses on limiting dissemination of information by self-interestedcompanies about possible uses that it has not evaluated. This focus arises out ofconcern that such communications, although potentially helpful, might not fairlyrepresent reliable scientific information because of a potential to mislead, given thatthe FDA has not had any opportunity to verify whether the communication isscientifically supported.

The FDA currently regulates off-label use of drugs through a complex and indirectscheme focused on companies;74 this regulation incentivizes companies to conductscientific studies to establish whether new uses are supported."5 There is no expresslegal prohibition of off-label promotion under the existing statute.76 Rather,

69. E.g., United States v. Caronia, 703 F.3d 149, 153 (2d Cir. 2012).70. This is consistent with the fact that the FDA evaluates whether a drug is safe and

effective for a stated indication. See supra Section I.A. In addition, companies can and dooften obtain approvals for many different uses of a drug. For example, Abbvie has obtainedmany approvals for different uses of its blockbuster drug Humira. IOth New Indication forHumira Approved by FDA, PHARMALETTER (Jan. 7, 2016) https://www.thepharmaletter.com/article/Oth-new-indication-for-humira-approved-by-fda [https ://perma.cc/CLG7-VUC6].

71. 21 U.S.C. § 396 (2012).72. Klasmeier & Redish, supra note 5, at 321-22.73. Id. at 322-23.74. See AmarinPharma, Inc. v. FDA, 119 F. Supp. 3d 196,205 (S.D.N.Y. 2015); see also

Nathan Cortez, The Statutory Case Against Off-Label Promotion, 83 U. CHI. L. REV. ONLINE

124 (2016) (explaining why FDA regulation of off-label promotion is justified).75. FDA Memorandum, supra note 6.76. See United States v. Caronia, 703 F.3d 149, 154 (2d Cir. 2012) ("The FDCA and its

accompanying regulations do not expressly prohibit the 'promotion' or 'marketing' of drugsfor off-label use."). However, labeling that suggests an unapproved use may cause a drug tobe "new," for that use, such that a company could be liable of introducing a new drug intointerstate commerce without approval. 21 C.F.R. § 202.1 (e)(4)(i)(a) (2017).

2019]

INDIANA LAW JOURNAL

companies, and their representatives, may be subject to criminal liability if theyintroduce, or conspire to introduce, a "misbranded" drug into interstate commerce.77

What constitutes a "misbranded" drug, in turn, includes if its "labeling" is inadequatefor "intended use."'78 Labeling is defined to include not just the actual label on a drug,but also accompanying documents, even if separately distributed.79 Intended use canbe established by advertising as well as oral or written statements by representativesof a company.80 Given the complex web of statutory provisions involved, as well asFirst Amendment challenges, the FDA has issued a series of guidance documentsover the years concerning what actions would immunize a company fromprosecution for misbranding.81

Misbranding is a serious issue for companies not only because it may result incriminal liability. In particular, a criminal conviction against a company due tomisbranding can result in the complete exclusion of all of its drugs from participationin federal health care programs, such as Medicare and Medicaid, which serve assignificant sources of revenue for most pharmaceutical companies.81 In addition, acompany that promotes a drug for off-label use may face substantial fines under theFalse Claims Act based on the theory that marketing of off-label use caused falseclaims to be submitted for government programs, such as Medicare, knowing thatsuch claims cannot be properly reimbursed.83

77. 21 U.S.C. § 331(a) (2012).78. 21 U.S.C. § 352 (a), (f) (2012).79. 21 U.S.C. § 321(m) (2012); 21 C.F.R. § 202.1(1)(2) (2017).80. 21 C.F.R. § 201.128 (2017). In addition, a recent revision added a "totality of

evidence" requirement. 82 Fed. Reg. 2193 (Jan. 9, 2017) (revising the standard of intendeduse for 21 C.F.R. § 201).

81. Thus far, guidance documents have increasingly liberalized what companies may do,especially in light of recent commercial speech cases. Indeed, the most recent guidancedocuments issued in June 2018 are a further expansion of the 2017 proposed guidelines.Compare U. S. DEP'T OF HEALTH & HUMAN SERVS., FOOD & DRUG ADMIN., CTR. FOR DRUG

EVALUATION & RESEARCH, CTR. FOR BIOLOGICS EVALUATION & RESEARCH, CTR. FOR DEVICES

& RADIOLOGICAL HEALTH & OFFICE OF THE COMM'R, DRUG AND DEVICE MANUFACTURER

COMMUNICATIONS WITH PAYORS, FORMULARY COMMITTEES, AND SIMILAR ENTITIES-

QUESTION AND ANSWERS: GUIDANCE FOR INDUSTRY AND REVIEW STAFF (2018), with U.S.DEP'T OF HEALTH & HUMAN SERVS., FOOD & DRUG ADMIN., CTR. FOR DRUG EVALUATION &

RESEARCH, CTR. FOR BIOLOGICS EVALUATION & RESEARCH, CTR. FOR DEVICES &

RADIOLOGICAL HEALTH & OFFICE OF THE COMM'R, DRUG AND DEVICE MANUFACTURER

COMMUNICATIONS WITH PAYORS, FORMULARY COMMITTEES, AND SIMILAR ENTITIES-

QUESTION AND ANSWERS: GUIDANCE FOR INDUSTRY AND REVIEW STAFF (2017); see also ROPES

& GRAY, FDA ISSUES FINAL GUIDANCE DOCUMENTS RELATING TO MEDICAL PRODUCT

MANUFACTURER COMMUNICATIONS (2018) (referring to guidance documents as providingmore flexibility to manufacturers, as well as noting specific changes such as expanding thesafe harbor to include new uses in addition to investigational products).

82. See 42 U.S.C. § 1320a-7b(f) (2012); see also Katrice Bridges Copeland, EnforcingIntegrity, 87 IND. L.J. 1033 (2012).

83. 18 U.S.C. § 287 (2012); 31 U.S.C. § 3729 (2012); see also Joan H. Krause, Truth,Falsity, and Fraud: Off-Label Drug Settlements and the Future of the Civil False Claims Act,71 FOOD & DRUG L.J. 401 (2016).

[Vol. 94:773


Despite the serious legal implications of off-label promotion, companies may beacting rationally in accepting these risks and promoting off-label drugs. Companiesseem to consider possible fines simply as part of the cost of doing business given thatvirtually all major pharmaceutical companies have settled for eye-popping amountsyet continue to promote drugs off-label.s4 For example, Pfizer paid $2.3 billion forillegally marketing the painkiller sold as Bextra, which was a record amount in 2009,yet it constituted less than three weeks of sales.s5 In addition, although the financialimplications of being barred from Medicare and Medicaid is a substantial threat,because the federal government has never barred a company and instead has settledsuits for a fraction of possible profits, it has not been a serious one.s6 Companies alsohave an incentive to promote drugs for all possible uses-even if not legallypermitted-due to the fact that such drugs are oftentimes patented but with arelatively short window for companies to recoup profits. In particular, a patent on adrug permits a company to exclude all others from making the identical drug duringthe effective patent term, which is roughly ten years after FDA approval.S" Moreover,since not all drugs investigated by a company make it to the marketplace, companieshave an incentive to maximize profits for FDA-approved patented drugs before thepatent expires and other companies can make identical versions (i.e., low-costgenerics) which can substantially undermine profits. Although a company could seekapproval for a use currently not approved, that takes time and money, and may notbe considered worthwhile if the patent term has expired or is soon to expire.

Although off-label use of drugs is not a new phenomenon, and in some situationsis reasonable,8 recent expansion of commercial speech to permit greater promotion

84. See, e.g., SAMMY ALMASHAT, SIDNEY M. WOLFE & MICHAEL CAROME, TWENTY-FIVEYEARS OF PHARMACEUTICAL INDUSTRY CRIMINAL AND CIVIL PENALTIES: 1991 THROUGH 2015

(2016), https://www.citizen.org/sites/default/files/231 1_0.pdf [http://perma.cc/3B25-73BY];Lena Groeger, Big Pharma's Big Fines, PROPUBLICA (Feb. 24, 2014), http://projects.propublica.org/graphics/bigpharma [https://perma.cc/K5X4-T6HS]; see also Aaron S.Kesslehim, Devan Darby, David M. Studdert, Robert Glynn, Raisa Levin & Jerry Avorn, FalseClaims Act Prosecution Did Not Deter Off-Label Drug Use in the Case of Neurontin, 30HEALTH AFF. 2318 (2011).

85. Gardiner Harris, Pfizer Pays $2.3 Billion to Settle Marketing Case, N.Y. TIMES (Sept.2, 2009), http://www.nytimes.com/2009/09/03/business/03health.htnml [https://perma.cc/26NK-AUC5].

86. E.g., Copeland, supra note 82, at 1035.87. 35 U.S.C. § 271 (2012). Althoughthe patenttermis longerthanthis, since a patented

drug can't be sold without FDA approval, which usually comes several years after the patentis granted, the commercially effective term is shorter. Henry G. Grabowski & John M. Vernon,Effective Patent Life in Pharmaceuticals, 19 INT'L J. TECH. MGMT. 98, 109 (2000) (findingmean of 11.7 years for analyzing drugs approved between 1990 and 1995). However, recentstudies suggest that due to patent protection of multiple aspects of a marketed drug, thecombined effective patent term could be longer. Amy Kapczynski, Chan Park & BhavenSampat, Polymorphs and Prodrugs and Salts (Oh My!.: An Empirical Analysis of"Secondary" Pharmaceutical Patents, PLOS ONE, Dec. 2012, at 6 (finding secondary patentscan add four to eleven additional years of patent term).

88. E.g., Amarin Pharma, Inc. v. FDA, 119 F. Supp. 3d. 196, 201-03 (S.D.N.Y. 2015)(noting that off-label uses are widely recognized, including by the FDA itself). This may betrue where there is no FDA approved treatment for a condition. See FDA Memorandum, supra

2019]

INDIANA LAW JOURNAL

of off-label uses creates a public health concern.9 Studies indicate that about threequarters of off-label uses have little or no scientific evidence to support them.9 Thiscan be problematic since even before the expansion of commercial speech for off-label marketing, one study suggested that approximately twenty percent ofprescriptions were for off-label uses.91 Similarly, before the commercial speechexpansion, there have been documented instances of death or serious injuriesincluding heart problems92 and strokes,93 based on off-label marketing.94 Fatalreactions from off-label uses of drugs may even result in the FDA adding warningsto the packaging even though the off-label use is not officially FDA regulated.95

note 6. Similarly, since most drugs are tested on adults, rather than children, off-label usage isalso important in that context. Alicia T. F. Bazzano, Rita Mangione-Smith, Matthias Schonlau,Marika J. Suttorp & Robert H. Brook, Off-Label Prescribing to Children in the United StatesOutpatient Setting, 9 ACAD. PEDIATRICS 81 (2009); Aaron S. Kesselheim & Jerry Avorn,PharmaceuticalPromotion of Physicians and FirstAmendmentRights, 358 NEw ENG. J. MED.1727, 1730 (2008).

89. In particular, off-label use can be the norm where there are no alternative treatments,such as some types of cancer. Similarly, off-label use is common for children since most drugsare approved for use only on adults. Accordingly, the FDA has stated that off-label uses maybe important or even constitute medically recognized standard of care in some situations. FDAMemorandum, supra note 6, at 17.

90. See Tewodros Eguale, David L. Buckeridge, Nancy E. Winslade, Andrea Benedetti,James A. Hanley & Robyn Tamblyn, Drug, Patient, and Physician Characteristics Associatedwith Off-Label Prescribing in Primary Care, 172 J. AM. MED. ASS'N: INTERNAL MED. 781,783 (2012) (finding that seventy-nine percent of off-label prescriptions in Canada lackedstrong scientific basis based on actual prescription data); David C. Radley, Stan N. Finkelstein& Randall S. Stafford, Off-Label Prescribing Among Office-Based Physicians, 166 J. AM.MED. ASS'N: INTERNAL MED 1021, 1025 (2006).

91. Randall S. Stafford, Regulating Off-Label Drug Use Rethinking the Role of the FDA,358 NEW ENG. J. MED. 1427, 1427 (2008).

92. HEALTH ACTION INT'L, supra note 41, at 29 (noting deaths from benfluorex, sold asMediator when used off-label as appetite suppressant); Kate Cohen, Fen Phen Nation, PBS(Nov. 13, 2003), https://www.pbs.org/wgbh/pages/frontline/shows/prescription/hazard/fenphen.html [https://perma.cc/S4N9-V6S4] (noting that the FDA had to withdraw drugscommonly combined as "Fen Phen" from the market in 1997 after studies by the Mayo Clinicrevealed the unapproved combination led to heart valve disease).

93. For example, Risperdal was marketed to doctors for elderly patients and used off labelfor dementia but can cause stroke-like attacks. Two years after such use, the manufacturerfinally sent a warning to doctors regarding this problem after Canadian drug regulators issueda warning and only after FDA officials had enough information to compel the company toissue the warning. Alison Young & Chris Adams, Prescribing Drugs "Off-Label" Is Routine,but Can Injure, Kill Patients, MCCLATCHY DC BUREAU (Nov. 2, 2003, 6:32 PM),https ://www.mcclatchydc.com/news/politics-govermnent/article24476629.html [https://perma.cc/74AP-KJG7] (noting that a patient suffered stroke-like heart attacks while usingRisperdal off-label).

94. In addition, sometimes off-label use puts patients at increased risk for serious diseases,such as cancer or Alzheimer's. This was the case with off-label prescribing of hormonereplacements to address menopause symptoms such as hot flashes. Nelson et al., supra note37, at 110-11 (explaining how hormone replacement is associated with risk of cancer).

95. For example, Rituxan was approved for non-Hodgkin's lymphoma, but when off-label prescriptions for avariety of unapproved uses led to severe and sometimes fatal reactions,

[Vol. 94:773


Notably, a recent study showed that adverse drug effects are more likely with off-label use, especially for those without strong scientific evidence, which constitutesthe majority of such uses.96 Finally, even if off-label use does not affirmatively harmpatients, it may still be a poor use of drugs for essentially ineffective treatments andresult in waste of resources. One prominent example involves Tamiflu, which waspromoted off-label to reduce complications based on industry-financed articles;many governments stockpiled Tamiflu based on off-label promotion, as well asselective publication of only two of ten studies made available to the FDA. 97

D. Introduction to Commercial Speech Law and Policy

The extent to which the FDA can regulate corporate speech surrounding off-labeluse of drugs fundamentally raises commercial speech issues under the FirstAmendment. Accordingly, this Section first explains what constitutes commercialspeech. Then, this Section explains fundamental distinctions between howcommercial and noncommercial speech are treated, as well as the reasons for thedistinctions. Finally, this Section explains the test that has historically been used toassess whether government regulation of commercial speech is consistent with theFirst Amendment. This Section is intended to introduce the general issues, whereaslater sections address how recent courts have addressed commercial speech thatimpacts off-label marketing more specifically.

An initial question is what constitutes commercial speech, since the governmentgenerally has more ability to regulate commercial speech.98 There is more than oneway that speech can be commercial, and thus subject to some governmentalrestrictions. Traditionally, speech has been considered commercial pursuant to athree-part test: (1) if it constitutes an advertisement, (2) refers to a specific product,or (3) is made with an economic motivation.99 Speech that does all three of these isdefinitively commercial; drug company distribution of scientific articles regarding

the FDA issued alerts and then a warning to the label. Rituximab (Marketed As Rituxan)Information, FOOD & DRUG. ADMIN., https://www.fda.gov/Drugs/DrugSafety/ucml09106.htm [https://perma.cc/PP5M-DYSM]; see also Rosoff & Coleman, supra note 23, at 669-70n.71.

96. Tewodros Eguale, David L. Buckeridge, Aman Verma, Nancy E. Winslade, AndreaBenedetti, James A. Hanley & Robyn Tamlyn, Association of Off-Label Drug Use andAdverse Drug Events in an Adult Population, 176 J. AM. MED. Ass'N: INTERNAL MED. 55(2016); see also Aaron S. Kesselheim, John Connolly, James Rogers & Jerry Avom,Mandatory Disclaimers on Dietary Supplements Do Not Reliably Communicate the IntendedIssues, 34 HEALTH AFF. 438, 438 (2015) [hereinafter Kesselheim et al., MandatoryDisclaimers] (noting poor efficacy or harm in recent years with promotion of off-label uses).

97. E.g., HEALTH ACTION INT'L, supra note 41; Brownlee & Lenzer, supra note 38 (notingonly two of ten studies published). This may have been an extreme case where the off-labeluse was suggested by the CDC, which has subsequently been revealed to have relied onindustry funding. Lenzer, supra note 38.

98. Va. State Bd. of Pharmacy v. Va. Citizens Consumer Council, Inc., 425 U.S. 748, 781(1976). However, before then, the Court had considered there to be no Constitutional restrainton "purely commercial advertising" that was well within the realm of "legislative judgment."Valentine v. Chrestensen, 316 U.S. 52, 54 (1942).

99. Va. State Bd. of Pharmacy, 426 U.S. at 761, 781.

2019]

INDIANA LAW JOURNAL

off-label uses of their drugs to doctors has been found to meet all three. 100 In addition,what is considered commercial speech has expanded in recent years. 101

First Amendment jurisprudence has traditionally treated commercial speechdifferently than noncommercial speech. Noncommercial speech that is traditionallyconsidered at the core of the First Amendment, such as political or religious speech,is generally immune from government restriction since restrictions are invalid unlessthey pass strict scrutiny. 102 Commercial speech, on the other hand, has traditionallybeen evaluated under a more lenient intermediate scrutiny test, as explained below. 103

The differing levels of scrutiny have generally resulted in individuals having a rightto freely express opinions, including false ones, whereas governments havetraditionally been permitted to regulate clearly false and misleading commercialstatements.

The reason for the difference relates to the underlying policy distinctions. Inparticular, all speech of individuals is considered to promote a "marketplace of ideas"that fosters better-informed citizenry, and is consistent with democratic self-determination;104 the democratic tie-in is important to permitting any individual tofreely express opinions, including ones that are false.105 This rationale also explainsthe few situations where individual speech is not protected, including fighting wordsand obscenity, which do not involve reasoned and deliberative response. 106 Althougha "marketplace of ideas" has been referenced in the context of commercial speech, itis generally only in the context of making intelligent decisions based on the free flowof commercial information without any tie to democratic principles. 10 7

Notably, commercial speech has historically only been protectable under the FirstAmendment because the information is valuable to the audience, but not because thespeaker has any right to speak.08 This policy distinction is why the government can

100. Wash. Legal Found. v. Friedman, 13 F. Supp. 2d 51, 64 (D.D.C. 1998).101. See Sorrell v. IMS Health Inc., 564 U.S. 552, 568-71 (2011); Citizens United v. FEC,

558 U.S. 310 (2010).102. Of course, this does not apply to speech that has no informational value, such as

obscenity or fighting words. See infra note 106.103. Sorrell obviously suggests that commercial speech regulation is subject to heightened

scrutiny based on viewpoint targeting, although what this means is still unclear. 564 U.S. 552.104. See id. at 583.105. See United States v. Alvarez, 567 U.S. 709, 718 (2012); Jennifer M. Keighley, Can

You Handle the Truth? Compelled Commercial Speech and the First Amendment, 15 U. PA. J.CONST. L. 539, 551 (2012).

106. Miller v. California, 413 U.S. 15, 24 (1973) (denying First Amendment protection tosexual materials with no "literary, artistic, political, or scientific value"); Brandenburg v. Ohio,395 U.S. 444 (1969) (barring speech that is likely to incite imminent lawless action);Chaplinksy v. New Hampshire, 315 U.S. 568, 571-72 (1942) (noting that profane, libelous,and 'fighting" words "are no essential part of any exposition of ideas, and are of such slightsocial value as a step to truth that any benefit that may be derived from them is clearlyoutweighed by the social interest in order and morality").

107. See, e.g., Va. State Bd. of Pharmacy v. Va. Citizens Council, Inc., 425 U.S. 748, 760(1976).

108. E.g., Ohralik v. Ohio State Bar Ass'n, 436 U.S. 447, 456 (1978) (noting that the stateright to regulate some commercial speech is because commercial speech is not a speaker-oriented autonomy right and thus gets more limited protection); Keighley, supra note 105, at

[Vol. 94:773


sometimes regulate commercial speech and, even in some cases, mandate thatcommercial speakers provide information to consumers under some situations basedon the same principle that commercial speech should inform consumers. 10 9 Althoughone scholar has argued First Amendment law, including commercial speech,recognizes speaker rights,11 courts have thus far not explicitly noted any reason todepart from the traditional audience -centric rationale for commercial speech.111

The traditional framework for analyzing whether regulation of commercial speechis permissible is according to the intermediate scrutiny test first announced by theSupreme Court's 1980 decision in Central Hudson. 112 More recently, the SupremeCourt majority in Sorrell v. IMS Health Inc. applied the Central Hudson test113 eventhough it controversially suggested that a regulation targeting commercial speech ofpharmaceutical companies should be subject to stricter scrutiny.114 The Sorrell

554; Robert Post & Amanda Shanor, Adam Smith 's FirstAmendment, 128 HARv. L. REv. 165,170, 172 (2015); Shanor, supra note 8, at 146 (noting that commercial speech doctrine wascreated as a tool of consumer protection during the consumer protection movement, and notto promote autonomy interests of commercial speakers). In addition, many believe that thedoctrine of free speech generally is intended to focus on right of audiences to access ideaswithout governmental interference, rather than the rights of speakers. E.g., Daniel A. Farber,Free Speech Without Romance: Public Choice and the First Amendment, 105 HARV. L. REV.554, 558-59 (1991); Eugene Volokh, The Freedom of Speech and Bad Purposes, 63 UCLAL. REv. 1366, 1370-71 (2016).

109. Zauderer v. Office of Disciplinary Counsel, 471 U.S. 626 (1985). However, thegovernment can only compel speech that is purely factual and noncontroversial, which hasrecently been at issue with FDA regulations that aimed to require mandatory images ontobacco labeling. As one court noted, although the images are not "patently false," they canalso be misunderstood or intended to evoke an emotional response. R.J. Reynolds TobaccoCo. v. FDA, 696 F.3d 1205, 1216-17 (D.C. Cir. 2012); see also Dist. Tobacco City & Lottery,Inc. v. United States, 674 F.3d 509, 560-61 (6th Cir. 2012) (finding that pictures are inherentlypersuasive).

110. Leslie Kendrick, Are Speech Rights for Speakers, 103 VA. L. REv. 1767 (2017).111. One commentator suggests Sorrell's majority was wrong to ignore the value, or,

rather, lack of value of detailing information for doctors. Alan B. Morrison, No Regrets(Almost): After Virginia Board of Pharmacy, 25 WM. & MARY BIL OFRTS. J. 949, 959 (2017).

112. Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm'n, 447 U.S. 557 (1980).113. 564 U.S. 552 (2011). Since the government regulation did not pass the intermediate

test, arguably there was no need to consider whether it would pass a heightened test. Id. at583-84.

114. Id. at 564-67. However, this is contrary to traditional commercial speech law thatgenerally permits direct regulation of content, even though this is not permissible regardingspeech of individuals. Cent. Hudson, 447 U.S. at 564 n.6. As some have noted, commercialspeech regulation is, by its very nature, content-based, such that this suggested standard wouldthreaten widely accepted regulatory activity. See Sorrell, 564 U.S. at 589 (Breyer, J.,dissenting) (noting that "regulatory programs necessarily draw distinctions on the basis ofcontent" and that it is not unusual for some regulations to be "speaker-based" affecting only aclass of entities, i.e. the regulated ones); Tamara R. Piety, "A Necessary Cost of Freedom "?The Incoherence of Sorrell v. IMS, 64 ALA. L. REv. 1, 5 (2012) ("Sorrell's reasoningeviscerates the rationale on which Virginia Pharmacy was based-protection of listeners'interests-and substitutes for it a rationale which elevates the interests of commercial speakersover that of listeners, such that even where the speech presents a detriment to listeners, it is

2019]

INDIANA LAW JOURNAL

majority opinion did not articulate a specific test and subsequent courts havesimilarly applied the same Central Hudson test.115 Accordingly, that remains thecurrent framework for analysis. There are four parts to analyze under Central Hudsonin terms of whether expression can be properly regulated by the federal or stategovernment. For commercial speech to be potentially protected, it must first concernlawful activity and not be actually misleading. Second, the government interest inregulating the speech must be substantial. Assuming both of those are satisfied, thethird and fourth questions are whether the government regulation directly advancesthat interest, and whether it is more extensive than necessary to serve that interest.

The first aspect of the Central Hudson inquiry permits the government to entirelyban commercial speech that is false or misleading. There is no First Amendmentprotection for such speech because there is no information value in inaccurate speechthat is more likely to deceive than inform. 116 Although the Supreme Court has framedthis inquiry around whether information is false or misleading, the only type ofinformation that is absolutely barred is inherently, rather than only potentially, falseor misleading information. The government has the burden to establish inherentlymisleading information, which is a high standard. This tough standard has beenrationalized as applicable because the government would otherwise get a "free pass"without considering other Central Hudson factors.117 On more than one occasion,courts have been reluctant to accept FDA assertions that the relevant audience willbe misled without empirical support of such confusion. 118

If information is truthful, but potentially misleading, government regulationdepends upon the three other factors. In particular, the government must have a"substantial interest" in the regulation, the regulation must "directly" advance the

protected because of its value to the speaker." (emphasis in original)); Shanor, supra note 8,at 146 (noting that commercial speech is content based).

115. See, e.g., Amarin Pharma, Inc. v. FDA, 119 F. Supp. 3d 196 (S.D.N.Y. 2015).Caronia's majority stated that strict scrutiny applied, but actually applied the Central Hudsontest. United States v. Caronia, 703 F.3d 149, 164-65 (2d Cir. 2012).

116. Cent. Hudson, 447 U.S. at 563-64.117. E.g., Rodney A. Smolla, SMOLLA & NIMMER, COMMERCIAL SPEECH 20:15 (2017

Update) (considering it "drastic medicine" to treat a statement as inherently misleadingbecause that would give regulators a "free pass" from the usual requirement of proof andcausation" and suggesting suggesting courts treat claims with a "healthy dose of skepticism").

118. This was true in a number of cases involving marketing of dietary supplements toconsumers. E.g., Int'l Dairy Foods Ass'n v. Boggs, 622 F.3d 628, 637 (6th Cir. 2010)(concluding that proposed claim informs consumers of a "meaningful distinction," contrary toFDA position and "at worst ... misleads" them to believe that milk untreated with rbST harmstheir health); Pearson v. Shalala, 164 F.3d 650, 655 (D.C. Cir. 1999) (dismissing FDAconcerns of consumer confusion and analogizing FDA presumptions concerning consumersas based on mistaken presumption that "consumers were asked to buy something whilehypnotized, and therefore they are bound to be misled"). In fact, one court even asserted thatit would not be inclined to find statements to be inherently misleading without proof of twodifferent things: minimal factual support for the allegedly false statement and empiricalsupport that consumers would be confused. Whitaker v. Thompson, 248 F. Supp. 2d 1, 10(D.D.C. 2002) (citing Pearson, 164 F.3d at 659-60). Although there is an argument thatconsumers may be more likely confused by marketing than doctors, the same burden of proofwould apply to an argument that doctors are confused.

[Vol. 94:773


government interest, and the regulation must be no more extensive than necessary.11 9

In other words, even if the government has a substantial interest in the regulation, acomplete ban might not withstand First Amendment scrutiny if it is considered morerestrictive than necessary to serve that interest.

In many situations involving government interest regarding regulation ofcommercial speech relating to foods and drugs, courts have overruled governmentregulations, finding them more extensive than necessary. For example, Thompson v.Western States, the first Supreme Court commercial speech case involving drugs,held in a 5-4 decision that specialty compounding pharmacies were permitted toadvertise their products (and could not be banned from all advertising) because therewere other nonspeech restrictions that could be used to satisfy the governmentinterest in discouraging large-scale manufacturing by these pharmacies, such asforbidding or limiting compounded drug sales.12' Also, in Washington LegalFoundation v. Friedman, the District Court for the District of Columbia similarlyfound prior FDA regulations regarding distribution of scientific articles and industry-sponsored education relating to off-label use to be overly restrictive on speech.121

This was based on an assumption that providing complete disclosure of themanufacturer interest would be a less burdensome restriction.122 Along similar lines,courts have invalidated FDA regulations that attempt to bar health claims for dietarysupplements without significant scientific support; although the courts recognized asubstantial interest in protecting consumers from deceptive marketing practices, thecourts found a complete bar more extensive than necessary without empirical proofthat a disclaimer would fail to correct deceptiveness. 123

II. REVEALING AND DEBUNKING SCHEMAS THAT SUPPORT FIRST

AMENDMENT FALLACIES

This Part explains the existence of two distinct schemas that courts haveperpetuated in cases expanding commercial speech protection for pharmaceuticalmarketing: the sophisticated doctor schema and the more information schema. Theseschemas are not intended to entirely account for all of commercial speechjurisprudence. Instead, the goal is to show that some legal presumptions reflect theseschemas rather than empirically supported fact. Understanding these schemas helpsshed light on recent First Amendment cases discussed in Part III, and also serves as

119. Cent. Hudson, 447 U.S. at 566.120. Thompson v. W. States Med. Ctr., 535 U.S. 357 (2002). The relatively lax oversight

of drug compounding has failed to promote public health. For example, in 2012, over 60people were killed and over 700 suffered from meningitis when a company shippedcontaminated drugs. Doris Browne, Fixing a Crisis of Confidence in Drug Compounding,WASH. EXAMINER (Jan. 16, 2018, 7:00 AM), https://www.washingtonexamiiner.com/fixing-a-crisis-of-confidence-in-drug-compounding/article/2645969 [https://perma.cc/2QGL-BECU].

121. 13 F. Supp. 2d 51, 72-73 (D.D.C. 1998).122. Id.123. Pearson, 164 F.3d. 650; Whitaker, 248 F. Supp. 2d 1. However, in the case of dietary

supplements, it has been noted that the potential for deception is "severely limited," in that noserious physical harm will result from consumption of such supplements. Whitaker, 248 F.Supp. 2d at 16.

2019]

INDIANA LAW JOURNAL

an important foundation for proposed changes suggested in Part IV. This Part beginswith an introduction to schemas based on the literature on cognitive biases. Then, theexistence of each schema in the case law is discussed before explaining why theschema is not completely supported by fact.

A. Introduction to Schemas

This Section provides fundamental information concerning key aspects from thevast areas of cognitive bias and behavioral economics that is essential tounderstanding the existence of two key schemas introduced in the next two sections.Essentially, researchers in these areas have repeatedly shown that all individuals,regardless of education or experience, acquire and process information subject tocertain largely subconscious mental flaws. In particular, although individuals believethat they approach information systematically and with deductive reasoning, researchshows this is often not the case due to reliance on cognitive biases, heuristics, andschemas, which are essentially mental shortcuts. 124 Incomplete cognitive processingis actually considered adaptive to deal with an information-rich world and especiallyto help efficiently deal with ambiguous situations and/or time-limited situations.125

On some level, individuals seem to intuit this since people often believe that othersare subject to bias-while simultaneously discounting the possibility of having theirown bias.126 Although all individuals are subject to cognitive biases, doctors may beespecially likely to use such processing not only because they are busy, but alsobecause the process of diagnosis often involves ambiguity and incomplete

124. See DANIEL KAHNEMAN, THINKING FAST AND SLOW (2011).125. In addition, although literature often refers to cognitive biases and the word "bias"

may suggest a flaw, they are important for psychological well-being. See Jeffrey J. Rachlinski,The Uncertain Psychological Case for Paternalism, 97 Nw. U. L. REv. 1165, 1172 (2003)(only individuals free of biased self-perceptions suffered from clinical depression); see alsoLEON FESTINGER, A THEORY OF COGNITIVE DISSONANCE 2-3 (1957) (noting that individualsare motivated to ignore or discredit information inconsistent with prior views to preventpsychological distress that would otherwise result; in other words, people try to avoid"cognitive dissonance").

126. This is referred to as nalve realism. Robert J. Robinson, Dacher Keltner, AndrewWard & Lee Ross, Actual Versus Assumed Differences in Construal: "Naive Realism" inIntergroup Perception and Conflict, 68 J. PERSONALITY & SOC. PSYCHOL. 404, 404 (1995);Brad J. Sagarin, Robert B. Cialdini, William E. Rice & Sherman B. Serna, Dispelling theIllusion of Invulnerability, 83 J. PERSONALITY & SOC. PSYCHOL. 526 (2002). For example, inthe 2012 presidential election, some Republicans assumed Nate Silver was biased, and thus,his data predicting an Obama win was suspect. Nate Silver, When Internal Polls Mislead, aWhole Campaign May Be to Blame, N.Y. TtES: FIVETHIRTYEIGHT (Dec. 1, 2012, 6:01 AM),http://fivethirtyeight.blogs.nytimes.com/2012/12/0 1/when-internal-polls-mislead-a-whole-campaign-may-be-to-blame [https://perma.cc/33NV-S4P7].

[Vol. 94:773


information, 127 which studies repeatedly show are situations where individuals areespecially likely to rely on cognitive biases. 128

1. Schemas Shape Information Processing

An important type of cognitive flaw in processing information for this Article isa schema. Essentially, a schema is a mental framework through which individualsperceive and filter information.129 Schemas are developed through concreteexperiences as well as personal morals and societal roles; some examples of schemasthat are part of popular culture include lawyers as mercenaries, women as shoppers,and men as relationship avoidant. Schemas operate like other cognitive flaws byoccuring largely outside of conscious awareness.

All individuals, including highly educated individuals such as doctors, scientists,and judges, hold schemas. Doctors are vulnerable to relying on schemas since theprocess of diagnosis typically involves great ambiguity, a situation in which schemasflourish.130 Studies indicate that doctors tend to generate an early hypothesis thatimpacts subsequent information gathering and may result in a faulty diagnosis."'Similarly, doctors may have schemas that impact how they prescribe drugs; theycould assume that new drugs are superior, or, alternatively, that new drugs pose morerisks. 132

Importantly, a schema impacts how new information is perceived. So, forexample, news that a huge iceberg has broken off from the Antarctic shelf is likelyperceived differently by those who believe in global warming, versus those who donot."' Schemas also impact how individuals perceive accuracy of information, as

127. See Mark A. Hall, Institutional Control of Physician Behavior: Legal Barriers toHealth Care Cost Containment, 137 U. PA. L. REV. 431, 480-82 (1988) (noting that, contraryto the assumption that doctors systematically evaluate possible treatments, they rely on generalrules or "heuristics" to deal with the fact that uncertainty pervades practices, given that manyprocedures have not been subject to rigorous clinical trials); Jessica Mantel, The Myth of theIndependent Physician: Implications for Health Law, Policy and Ethics, 64 CASE W. RES. L.REV. 455, 471 (2013); see also CONGRESSIONAL BUDGET OFFICE, RESEARCH ON THE

COMPARATIVE EFFECTIVENESS OF MEDICAL TREATMENTS 9 (2007) (noting that less than halfof medical decisions are supported by adequate evidence regarding effectiveness).

128. E.g., Dennis A. Gioia & Peter P. Poole, Scripts in Organizational Behavior, 9 ACAD.

MGMT. REV. 449, 454 (1984).129. AUGOUSTINOS ET AL., supra note 3, at 68-69.130. Mark P. Higgins & Mary P. Tully, Hospital Doctors and Their Schemas About

Appropriate Prescribing, 39 MED. EDUC. 184 (2005).131. Elke U. Weber, Ulf Bckenholt, Denis J. Hilton & Brian Wallace, Determinants of

Diagnostic Hypothesis Generation: Effects of Information, Base Rates, and Experience, 19 J.EXPERIMENTAL PSYCHOL. 1151 (1993).

132. See Cynthia M. Ho, Drugged Out. How Cognitive Bias Hurts Drug Innovation, 51SAN DIEGO L. REV. 419, 499-504 (2014); Peter R. Mansfield, Healthy Skepticism's NewAdWatch: Understanding Drug Promotion, 179 MED. J. AUST. 644 (2003) [hereinafterMansfield, Healthy Skepticism]; Clement J. McDonald, Medical Heuristics: The SilentAdjudicators of Clinical Practice, 124 ANNALS INTERNAL MED. 56, 57, 59 (1996).

133. Compare Ilissa Ocko, Huge Antarctic Iceberg Breaks Off Here's Why It WorriesScientists, ENVTL. DEF. FUND (July 12, 2017), https://www.edf.org/blog/2017/07/12/huge

2019]

INDIANA LAW JOURNAL

well as whether that information is persuasive. The same information will beprocessed differently by those with different schemas. More specifically, schemascan influence the extent to which an individual considers or discounts research resultsdepending on whether or not it is consistent with their schema.134

2. Schemas are Reinforced by Other Cognitive Biases

Schemas are notably resistant to change. According to research, the more peoplerely on a schema, the more the schema is resistant to any inconsistent evidence, sinceschemas dictate what information an individual will remember. 135 In other words,once a schema is established, individuals view new information consistent with thatschema and even seek out similar information. Schemas are often maintained by ahost of ancillary cognitive biases including confirmation bias, availability, andoverconfidence;136 studies have found doctors to rely on these and other cognitivebiases. 137

Confirmation bias is very important to maintaining a schema. This refers to thetendency to view information consistent with preconceptions and even skew searchesfor more information to be consistent with such preconceptions. 138 Essentially,

-antarctic-iceberg-breaks -here s-why-it-worries -scientists [https ://perma.cc/D6WH-NM32],with Dave Mosher, Scientists Say the Delaware-Size Iceberg that Broke OffAntarctica CouldHave Happened Without Global Warming, Bus. INSIDER (July 18, 2017, 3:29 PM), https://www.businessinsider.com/iceberg-a68-did-climate-change-cause-crack-2017-7 [https://perma.cc/2G9A-B64K]; see also Justin Worland, New Discovery in Antarctica Suggests IceSheets Could Disappear Way Faster than Previously Thought, TIME (Apr. 19, 2017),http://time.con/4745827/antarctica-water-climate-change [https://perma.cc/U8UJ-W5VP].

134. In a classic study, subjects were preselected based on their preference for or againstcapital punishment. Individuals were more critical of research methods inconsistent with theirbeliefs, citing issues such as insufficient sample size, nonrandom sample selection, or absenceof control for key variables. Charles G. Lord, Lee Ross & Mark R. Lepper, BiasedAssimilationand Attitude Polarization: The Effects of Prior Theories on Subsequently ConsideredEvidence, 37 J. PERSONALITY & SOC. PSYCHOL. 2098 (1979); see also Harold H. Kassarjian &Joel Benjamin Cohen, Cognitive Dissonance and Consumer Behavior, 8 CAL. MGMT. REv. 55(1965) (smokers were motivated to disregard evidence of health risks of smoking); ZivaKunda, The Case for Motivated Reasoning, 108 PSYCHOL. BULL. 480, 490 (1990) (showingsubjects with high caffeine consumption likely to disbelieve article that supposedly correlatedcaffeine with a serious disease and to consider the research less sound).

135. E.g., RICHARD E. NISBETT & LEE Ross, HUMAN INFERENCE: STRATEGIES AND

SHORTCOMINGS OF SOCIAL JUDGMENT 146-50 (1980).136. In addition, a schema may be also reinforced by "anchoring," a cognitive bias that

gives undue weight to initial information. Amos Tversky & Daniel Kahneman, JudgmentUnder Uncertainty: Heuristics and Biases, 185 Sci. 1124, 1128 (1974). So, once a schema isestablished, it may have an anchoring effect against which new information is processed.

137. E.g., Gustavo Saposnik, Donald Redelmeier, Christian C. Ruff& Philippe N. Tobler,Cognitive Biases Associated with Medical Decisions: A Systematic Review , 16 BIOMED CENT.138 (2016) (finding that all studies over a thirty-five year period indicate doctors are subjectto at least one cognitive bias and certain ones are associated with diagnostic errors, includingoverconfidence).

138. Internet searches can even feed into this since current algorithms are designed to tailorsearch results to prior searches.

[Vol. 94:773


people see what they want to see. So, individuals with different prior beliefs mayhave very different reactions to the same "facts" (i.e., they will be more likely toaccept new information similar to existing views and be skeptical to information thatis inconsistent with those views). Confirmation bias is most likely to play a role whentrying to comprehend ambiguous information that would be more susceptible todifferent interpretations.13 9 In such situations, individuals with opposing schemasmay become more rooted in those schemas after viewing the same "evidence.""14 Inaddition, even if the conclusion from our confirmation biases is objectivelyunsupported, the bias is difficult to combat.141 Studies have shown that even afterinformation is discredited an individual may continue to believe it; empirical studieshave shown this is the case with the myth that President Obama is not a U.S. citizen,and especially so among conservatives who would typically be more likely to hold anegative view of him. 142

Doctors are also subject to confirmation bias. Doctors may favor their initialhypotheses or assumptions due to confirmation bias. 143 For example, while somedoctors erroneously assumed most patients are resistant to aspirin as a low-costdiuretic, which some companies with more expensive drug alternatives to aspirin hadpromoted, this assumption was recently shown to have no scientific justification. 144

Doctors could have been influenced by company advertising to favor the self-interested corporate idea that patients are resistant to aspirin as a type of confirmationbias.

The creation, as well as propagation, of a schema can also be related to the"availability" bias. Essentially, this cognitive shortcut refers to individuals moreeasily recalling readily accessible information and assuming it is necessarilyimportant. In the medical context, the availability heuristic may explain why a 2005

139. One of the most striking examples is that researchers found that conservativesassumed that the political satire The Colbert Report, was in fact truthful because the showreflected their own political view. Heather L. LaMarre, Kristen D. Landreville & Michael A.Beam, The Irony of Satire: Political Ideology and the Motivation to See What You Want toSee in The Colbert Report, 14 INT'L J. PRESS/POL. 212, 222-23 (2009).

140. For example, believers and nonbelievers became more entrenched in their views afterreading a fictitious report concerning religious resurrection. C. Daniel Batson, RationalProcessing or Rationalization?: The Effect of Disconfirming Information on a StatedReligious Belief 32 J. PERSONALITY & SOC. PSYCHOL. 176, 176 (1975); see also Lord et al.,supra note 134 (finding similar results after subjects were provided "evidence" about thedeterrent effects of capital punishment).

141. E.g., George Loewenstein, Samuel Issacharoff, Colin Camerer & Linda Babcock,Self-Serving Assessments of Fairness and Pretrial Bargaining, 22 J. LEGAL STUD. 135, 139(1993).

142. E.g., Brendan Nyhan & Jason Reifler, When Corrections Fail: The Persistence ofPolitical Misperceptions, 32 POL. BEHAV. 303, 323 (2010).

143. See Weber et al., supra note 131.144. Tilo Grosser, Susanne Fries, John A. Lawson, Shiv C. Kapoor, Gregory R. Grant &

Garret A. FitzGerald, Drug Resistance and Pseudoresistance: An Unintended Consequence ofEnteric Coating Aspirin, 127 CIRCULATION 377, 382-83 (2013); Katie Thomas, Study RaisesQuestions on Coating of Aspirin, N.Y. TuMEs (Dec. 4, 2012), https://www.nytimes.com/20 12/12/05/busines s/coating-on-buffered-aspirin-may-hide -its-heart-protective -effects.html [https ://perma.cc/V2WH-CRNW].

2019]

INDIANA LAW JOURNAL

study found that doctors were undertreating patients' pain because they had concernsabout addiction to painkillers due to the high publicity of opiate addiction. 145 Theavailability heuristic can impact the strength of a schema. If information is moreavailable in the sense of being repeated, it is presumed correct.146 Moreover, therepetition of information is often remembered even better than its source, 117 withrepeated information often assumed to be from a credible source when the source isnot recalled.148 Repeated information is especially powerful to individuals that lackmotivation or opportunity to scrutinize the message.14 9 Notably, even whenindividuals are told that repeated statements are not more likely to be true or eventhat the source of the information is biased, they still seem inclined to assumerepeated statements are true.15 This is especially relevant to the context ofpharmaceutical marketing to doctors who realize that information from companies islikely to be biased. In particular, this data suggests that doctors repeatedly toldinformation from companies may assume that company information is true.

In addition, a schema may be supported by an overconfidence bias.1 Theoverconfidence bias refers to an individual's overestimation of likely being correctregarding abilities or knowledge.152 For example, the vast majority of entering lawstudents believe they will be in the top ten percent of the class when, statistically,that is impossible. The overconfidence bias applies to all individuals, including

145. See Jill G Klein, Five Pitfalls in Decisions About Diagnosis and Prescribing, 330BRIT. MED. J. 781, 782 (2005).

146. See KAHNEMAN, supra note 124, at 129-35.147. Danielle C. Polage, Making Up History: False Memories of Fake News Stories, 8

EUR. J. PSYCHOL. 245, 248-49 (2012).148. Alison R. Fragale & Chip Heath, Evolving Informational Credentials: The

(Aisgattribution of Believable Facts to Credible Sources, 30 PERSONALITY & SOC. PSYCHOL.

BULL. 225, 225-26 (2004); Kimberlee Weaver, Stephen M. Garcia, Norbert Schwarz & DaleT. Miller, Inferring the Popularity ofan Opinion from Its Familiarity: A Repetitive Voice CanSound Like a Chorus, 92 J. PERSONALITY& SOC. PSYCHOL. 821, 821-22 (2007).

149. Scott A. Hawkins & Stephen J. Hoch, Low Involvement Learning: Memory WithoutEvaluation, 19 J. CONSUMER RES. 212, 212-13 (1992); see also Sunita Sah, Don A. Moore &Robert J. MacCoun, Cheap Talk and Credibility: The Consequences of Confidence andAccuracy on Advisor Credibility and Persuasiveness, 121 ORGANIZATIONAL BEHAV. & HUM.DECISION PROCESSES 246 (2013) (individuals are strongly influenced by confident messagesand will not attempt to ascertain whether the message is correct if that requires effort).

150. Linda A. Henkel & Mark E. Mattson, Reading Is Believing: The Truth Effect andSource Credibility, 20 CONSCIOUSNESS & COGNITION 1705, 1705-06 (2011); see alsoFrederick T. Bacon, Credibility of Repeated Statements, 5 J. EXPERIMENTAL PSYCHOL.: HUM.LEARNING & MEMORY 241, 251 (1979) (showing that repeated statements are more likely tobe judged as true, whereas contrary statements are likely to be viewed as false).

151. Doctors likely also overestimate the effectiveness of patient treatments pursuant to anover optimism bias, pursuant to which an individual may overestimate likely positiveoutcomes. Tali Sharot, The Optimism Bias, 21 CURRENT BIOLOGY R941, R943 (2011). Thismay not perpetuate a schema, but could make doctors vulnerable to marketing.

152. Barbara Mellers & A. Peter McGraw, Self-Serving Beliefs and the Pleasure ofOutcomes, in 2 THE PSYCHOLOGY OF ECONOMIC DECISIONS: REASONS AND CHOICES 31, 37(Isabelle Brocas & Juan D. Carrillo eds., 2004); Rachlinski, supra note 125, at 1172-73.

[Vol. 94:773


doctors.153 A doctor or judge may thus be overly confident that his schema is correctand maintain it even in the face of contradictory evidence.

Although there are many individual cognitive biases, these examples shouldunderscore how schemas become entrenched and difficult to combat, even in the faceof contrary evidence. There is no comprehensive list of schemas, since they areinformed, in part, by individual experience. However, as discussed in the nextsections, there are two key schemas that have played an important role in commercialspeech jurisprudence regarding off-label marketing.

B. The Sophisticated Doctor Schema

This Section argues that there is an important, yet generally unrecognized, schemaoveremphasizing how sophisticated doctors are in evaluating information fromcompanies, including marketing information, as well as peer review publications.Although considering doctors more knowledgeable than lay consumers is consistentwith other areas of the law, such as the learned intermediary doctrine,154 thesophisticated doctor schema may result in courts unrealistically assuming thatdoctors are invulnerable to marketing influences. This Section first providesevidence of the schema in court opinions, as well as from scholarly literature anddoctors themselves. Then, this Section will explain why this schema is fundamentallynot empirically supported, even if it is a widespread presumption.

1. What Is the Sophisticated Doctor Schema?

In Washington Legal Foundation v. Friedman,155 the sophisticated doctor schemaplayed an important role in the first direct challenge to FDA regulation of off-labelpromotion. In this case, Washington Legal Foundation, an organization that focuseson challenging "undue influence from government" regulation,156 challenged twoFDA guidance documents, one of which intended to restrict the type of printedmaterials manufacturers could provide doctors concerning off-label uses and theother intending to limit manufacturer involvement in continuing medical

153. See Andrea 0. Baumann, Raisa B. Deber & Gail G. Thompson,Overconfidence Among Physicians and Nurses: The 'Micro-Certainty, Macro-Uncertainty'Phenomenon, 32 Soc. SCL MED. 167, 172-73 (1991) (doctors overconfident in "correct"treatment in situation where there was no empirically supported superior method of treatment);Franqois Laure, Sophie M. Colleau, Alain Fontaine & Louis Brasseur, Oncologists andPrimary Care Physicians' Attitudes Toward Pain Control and Morphine Prescribing inFrance, 76 CANCER 2375 (1995) (explaining that doctors are overly confident in the ability tocontrol pain).

154. See Sterling Drug Inc. v. Cornish, 370 F.2d 82, 85 (8th Cir. 1966). However, whetherthis doctrine is applicable has been challenged by one scholar. Kate Greenwood, PhysicianConflicts of Interest in Court: Beyond the "Independent Physician " Litigation Heuristic, 30GA. ST. U. L. REV. 759 (2014).

155. 13 F. Supp. 2d 51, 54 (D.D.C. 1988).156. Wash. Legal Found., 13 F. Supp. 2d at 54 (quoting from complaint); see also About

WLF, WASH. LEGAL FOUND., https://www.wlf.org/about-washington-legal-foundation/[https://perma.cc/C9CN-ENXC].

2019]

INDIANA LAW JOURNAL

education.15 7 The district court held that the FDA guidance documents wereunconstitutional violations of commercial speech rights158 despite the governmentinterest in encouraging supplemental applications for new drug uses because theregulation was viewed as more extensive than necessary. 15 9 The court recognizedthat since companies clearly want to increase sales,16 they would likely aggressivelypromote a single article supporting use of a drug-even if there is considerableevidence to the contrary-with a potential to mislead.161 Nonetheless, when it appliedthe Central Hudson test, the court seemed unconcerned about the potential to misleadphysicians. It stated that "despite the FDA's occasional statements in its briefs to thecontrary, physicians are a highly educated, professionally-trained and sophisticatedaudience" 162 in contrast with the "general consumer public." 16 3

The Washington Legal Foundation court seemed strongly influenced by thesophisticated doctor schema in rejecting out of hand the FDA's interest in ensuringdoctors have accurate and unbiased information. The court asserted that the FDA'sinterest in providing doctors complete and unbiased information was not a substantialgovernment interest that could support government regulation pursuant to theCentral Hudson test.164 The court asserted that doctors are "certainly capable ofcritically evaluating" information provided by self-interested companies.165 Thecourt claimed that it was "unclear" why a doctor might not critically evaluatescientific findings if presented by a company, even though earlier in the opinion thecourt noted that companies have an interest in selectively providing partialinformation to doctors. 166

An important part of the sophisticated doctor schema is the idea that doctors arecapable of evaluating articles provided by self-interested companies concerningwhether an unapproved use of a drug is effective, such that there is no need to relyon the FDA to assess whether drugs are safe and effective. For example, theWashington Legal Foundation court stated that "a physician's livelihood dependsupon the ability to make accurate, life-and-death decisions based upon the scientificevidence before them. They are certainly capable of critically evaluating journal

157. Final Guidance on Industry-Supported Scientific and Educational Activities, 62 Fed.Reg. 64074 (Dec. 3, 1997); Advertising and Promotion; Guidance, 61 Fed. Reg. 52800 (1996).

158. Wash. Legal Found., 13 F. Supp. 2d at 74 (finding guidance documents moreextensive than necessary, such that the fourth prong of the Central Hudson test fails); see alsoThompson v. W. States Med. Ctr., 535 U.S. 357, 374 (2002) (noting that it is a "questionableassumption that doctors would prescribe unnecessary medications," with regard to thepossibility that doctors might prescribe drugs that are advertised to consumers, in strikingdown a ban against advertising of compounded drugs).

159. Wash. LegalFound., 13 F. Supp. 2dat69.160. Id. at 64.161. Id. at65.162. Id. at 63 (emphasis added).163. Id. at 70 (" [M]anufacturers are not seeking to distribute this information to the general

consumer public, who likely lack the knowledge or sophistication necessary to make informedchoices on the efficacy of prescription drugs.").

164. Id. at 69-70.165. Id. at 70; see also id. at 69 (asserting that limiting material from doctors out of concern

for doctor misuse to be "wholly and completely unsupportable").166. Id. at 70.

[Vol. 94:773


articles or textbook reprints that are mailed to them, or the findings presented at CMEseminars." 167 The fact that doctors may be making life-and-death decisions based onsuch information does not mean that the incomplete information is not misleading.In addition, even if doctors realize the information is biased, and somewhat discountthe results as more favorable than necessary, doctors still lack complete information.

Since Washington Legal Foundation, the sophisticated doctor schema has beenembraced not only by other courts, as will be subsequently discussed in Part III, butalso by physicians. For example, some doctors participated as plaintiffs alongsidethe company Amarin that was seeking to promote off-label uses of its drugs in a suitchallenging FDA regulations limiting off-label promotion on First Amendmentgrounds.168 These doctors clearly do not share the FDA's concern that they may bemisled by drug manufacturers; rather, they assume that information from companieswill be valuable. 16 9

The views of the doctors in this suit are consistent with other studies indicatingthat doctors have a generally positive view regarding the most common type ofinformation from companies: pharmaceutical detailing (i.e., information fromindividual sales representative visits to doctors). Although some doctors recognizedetailing information may be biased or inaccurate, 170 many consider detailing to bea valuable source of information.171 Most studies indicate that doctors believe their

167. Id.168. AmarinPharma, Inc. v. FDA, 119 F. Supp. 3d 196, 199 (S.D.N.Y. 2015) (four medical

doctors joined with the pharmaceutical company Amarin in suing the FDA); see also Sorrellv. IMS Health Inc., 564 U.S. 552, 578 (2011) (noting that doctors find detailing "veryhelpful").

169. Amarin Pharma, 119 F. Supp. 3d at 213 (alleging that FDA rules operate "to keepdoctors ... in the dark about all the options for drug therapy they are legally empowered toprescribe" (quoting paragraphs 8-9 of the Complaint)).

170. See David Strang, Micheline Gagnon, William Molloy, Michel Bddard, PeterisDarzins, Edward Etchells & Warren Davidson, National Survey on the Attitudes of CanadianPhysicians Towards Drug-Detailing by Pharmaceutical Representatives, 29 ANNALS ROYAL

C. PHYSICIANS & SURGEONS CAN. 474 (1996) (finding that eighty percent of doctors agree thatdetailing overstates drug effectiveness). The reality is that detailing tends to be incomplete.See Barbara Mintzes, Joel Lexchin, JasonM. Sutherland, Marie Dominique Beaulieu, MichaelS. Wilkes, Genevi&ve Durrieu & Ellen Reynolds, Pharmaceutical Sales Representatives andPatient Safety: A Comparative Prospective Study of Information Quality in Canada, Franceand the United States, 28 J. GEN. INTERNAL MED. 1368, 1374 (2013).

171. E.g., Elisabeth H. Creyer & Ilias Hrsistodoulakis, Marketing PharmaceuticalProducts to Physicians: Sales Reps Influence Physicians Impressions of the Industry,MARKETING HEALTH SERVS., Summer 1998, 34, 36-38 (noting that sales representatives are akey source of information and that fifty-five percent of residents surveyed believedinformation was accurate, even though less than half considered representatives to betrustworthy); Lisa D. Spiller & Walter W. Wymer, Jr., Physicians' Perceptions and Uses ofCommercial Drug Information Sources: An Examination of Pharmaceutical Marketing toPhysicians, 19 HEALTH MARKETING Q. 91, 97 (2001) (finding that a vast majority ofphysicians considered the visits "somewhat helpful," with nearly twenty percent consideringthe visits "extremely helpful"); Michael G. Ziegler, Pauline Lew & Brian C. Singer, TheAccuracy of Drug Information Jrom Pharmaceutical Sales Representatives, 273 J. AM. MED.

ASS'N, 1296, 1298 (1995) (noting that seventy-four percent of residents found pharmaceuticalrepresentative information to be useful); see also HOUSE OF COMMONS: HEALTH CoMM., THE

2019]

INDIANA LAW JOURNAL

prescription behavior is not influenced by detailing and claim that they are able tocritically evaluate information from sales representatives;172 although some doctorsrealize that there may be some influence. 173 However, in general, doctors believeonly other doctors are susceptible. 174 This is an apt example of the overconfidence

INFLUENCE OF THE PHARMACEUTICAL INDUSTRY, 2004-05, HC 42-I, at 64 (UK) (criticizingdoctors for uncritically relying on industry information); Helen Prosser, Solomon Almond &Tom Walley, Influences on GPs'Decision to Prescribe New Drugs The Importance of WhoSays What, 20 FAm. PRAC. 61, 64 (2003) (noting that seventy percent of UK doctors consideredsales representatives an efficient way of obtaining information about drug information).

172. See Freek Fickweiler, Ward Fickweiler & Ewout Urbach, Interactions BetweenPhysicians and the Pharmaceutical Industry Generally and Sales Representatives Specificallyand Their Association with Physicians' Attitudes and Prescribing Habits: A SystematicReview, 7 BRIT. MED. J. OPEN 1, 3 (2017) (noting that a majority of studies found that doctorsdo not believe interactions with industry reps had any impact on prescription behavior);Melissa A. Fischer, Mary Ellen Keough, Joann L. Baril, Laura Saccoccio, Kathleen M. Mazor,Elissa Ladd, Ann Von Worley & Jerry H. Gurwitz, Prescribers and PharmaceuticalRepresentatives: Why Are We Still Meeting?, 24 J. GEN. INTERNAL MED. 795, 796 (2009)(finding that although most doctors recognized that there were studies indicating the influenceof detailing on prescriptions, they nonetheless believed that they were not unduly influenced,with some asserting that they were able to adequately glean useful information from themarketing). The same phenomena is seen with doctors in other countries. E.g., Prosser et al.,supra note 171, at 64 (noting that general practitioners in the UK felt able to separate crediblefrom misleading information, even though they generally questioned the objectivity of theindustry); Aldo De Ferrari, Cesar Gentille, Long Davalos, Leandro Huayanay & GermanMalaga, Attitudes and Relationship Between Physicians and the Pharmaceutical Industry in aPublic General Hospital in Lima, Peru, PLoS ONE, June 2014, at 3 (noting that almost ninetypercent of doctors in Peru assumed gifts and free lunches had no impact on prescribingbehavior); Mario Cdsar Scheffer, Interactions Between Pharmaceutical Companies andPhysicians Who Prescribe Antiretroviral Drugs for Treating AIDS, 132 SAO PAULO MED. J.55, 58 (2014) (noting that forty percent of doctors in Brazil claimed that interactions withrepresentatives had no influence whatsoever on their prescriptions).

173. Fickweiler et al., supra note 172, at 3 (noting studies where some doctors admitpossible influence of pharmaceutical detailing); see also Klaus Lieb & Armin Scheurich,Contact Between Doctors and the Pharmaceutical Industry, Their Perceptions and the Effectson Prescribing Habits, PLoS ONE, Oct. 2014, at 3 (finding forty-two percent of Germandoctors believed their prescribing habits were influenced).

174. John A. Hopper, Mark W. Speece & Joseph L. Musial, Effects of an EducationalIntervention on Residents' Knowledge and Attitudes Toward Interactions withPharmaceuticalRepresentatives, 12 J. GEN. INTERNAL MED. 639, 641 (1997) (doctors stronglydisagreed that interactions with pharma representatives would influence their own prescriptionbehavior, but assumed that most other doctors would be susceptible); M. A. Morgan, J. Dana,G. Loewenstein, S. Zinberg & J. Schulkin, Interactions of Doctors with the PharmaceuticalIndustry, 32 J. MED. ETiCs 559 (2006) (most OB-GYN doctors surveyed believed that otherswould be more likely to be influenced); Michael A. Steinman, Michael G. Shlipak & StephenJ. McPhee, OfPrinciples andPens, 110 AM. J. MED. 551 (2001) (finding that sixty-one percentof internal medicine residents claimed that promotions, such as gifts, had no impact on them,but only sixteen percent thought this was true of other doctors). However, how questions arephrased in studies may get different results. Stephen K. Sigworth, Mary D. Nettleman & GailM. Cohen, Pharmaceutical Branding of Resident Physicians, 286 J. AM MED. ASS'N 1024,1025 (2001) (finding over ninety percent of residents admitted that detailing had "some

[Vol. 94:773


bias in action, as studies indicate doctors are in fact influenced by pharmaceuticalmarketing, as discussed in the next section.

The sophisticated doctor schema is also consistent with the fact that doctorsstrongly value their clinical autonomy. In fact, the FDA's current circuitous attemptto regulate off-label use of drugs may be a reaction to physician opposition to moredirect regulation. For example, in the 1960s, the FDA was reportedly criticized bythe medical community when the FDA asked the author of a medical textbook towarn against unsafe dosage of a product that resulted in several deaths. 17 Physicianopposition to adding information about dangerous dosages of a product is consistentwith doctors holding a schema that they do not need information from someoneoutside their profession about dangers of a drug. This schema is consistent with ageneral culture of resisting perceived intrusions on independent professionaljudgments, with some suggesting that medical training and professional orientationmake them wary of ceding clinical discretion to a nonphysician.17 6 Some studies notethat initial medical training has a long-term impact and that disregard of evidence-based information may be a function of their experience during residency where theattending physician's opinion was considered all-important over other informationsources. 177

The sophisticated doctor schema may also result in some doctors valuing theirown clinical experience over data in peer review publications. Studies indicate thatdoctors consider their own clinical experience very important in deciding what toprescribe.17

1 Some studies indicate that doctors not only express disbelief aboutscientific studies that contradict their perceived clinical practice, but also express astrong belief that their clinical experience is superior.179 In other words, the

influence").175. Klasmeier & Redish, supra note 5, at 321.176. E.g., Mark A. Hall, Institutional Control of Physician Behavior: Legal Barriers to

Health Care Cost Containment, 137 U. PA. L. REV. 431, 451, 462-63 (1988); see also JamesL. Reinersten, Zen and the Art of Physician Autonomy Maintenance, 138 ANNALS INTERNALMED. 992 (2003) (noting common physician objections to guidelines).

177. Johnson, supra note 22, at 76-77 (citing Stefan Timmermans & Alison Angell,Evidence Based Medicine, Clinical Uncertainty, andLearning to Doctor, 42 J. HEALTH & SOC.BEHAV. 342, 345-47 (2001)).

178. Jerry Avom, Milton Chen & Robert Hartley, Scientific Versus Commercial Sourcesof Influence on Prescribing Behavior of Physicians, 73 AM. J. MED. 4 (1982) (noting thateighty-eight percent of doctors considered their own experience very important and moreimportant than scientific papers or other sources); see also Niteesh K. Choudhry, Robert H.Fletcher & Stephen B. Soumerai Systematic Review: The Relationship between ClinicalExperience and Quality of Health Care, 142 ANNALS OF INTERNAL MED. 260, 262-63 (2005)(finding that doctors relied on their own beliefs rather than empirical studies or guidelines byinstitutions such as the National Cancer Institute).

179. Rebecca K. Schwartz, Stephen B. Soumerai & Jerry Avorn, Physician Motivationsfor Nonscientific Drug Prescribing, 28 Soc. SCI. & MED. 577, 579 (1989) (noting that doctorsclaim that their " 'real-world medicine,' was more relevant to clinical practice than academicstudies"); see also Narcyz Ghinea, Ian Kerridge, Miles Little & Wendy Lipworth, Challengesto the Validity of Using Medicine Labels to Categorize Clinical Behavior: An Empirical andNormative Critique of "Off-Label" Prescribing, 23 J. EVALUATION CLINICAL PRAC. 574, 577(2017) (noting that doctors are skeptical that regulatory approval is more meaningful than their

2019]

INDIANA LAW JOURNAL

sophisticated doctor schema not only makes doctors believe they are invulnerable tocommercial marketing, but also that their experience is more valuable than scientificevidence. Understanding this helps to explain why doctors are resistant to changingtheir practices in light of scientific evidence. For example, one study shows that ittakes from ten to twenty years of evidence before there is widespread adoption ofscientifically supported clinical changes,18 and doctors are often resistant toadopting guidelines from organizations such as the National Cancer Institute or theNational Cholesterol Education Program.181 In extreme cases, doctors sometimescontinue with medical practices that studies have affirmatively shown to beinferior.182 This resistance could be a result of the same sophisticated doctor schema.Of course, for the purpose of this discussion, doctors in general are considered to allshare this trait, although there are some differences by age and gender. 183

In addition, legal academics may also indulge in the sophisticated doctor bias. Forexample, one First Amendment scholar argued that if doctors often prescribemedicine for off-label use

[i]t cannot be that doctors do not know what they are doing, do not carefor their patients, or have somehow been seduced or suckered by drugcompanies .... The most plausible intuitive answer.., is that doctors.• . have made the independent, professional medical judgment that the

own clinical opinion since countries are not consistent in approval of the same drugs). This isconsistent with the overconfidence cognitive bias.

180. E. A. Balas & S. A. Boren, Managing Clinical Knowledge for Health CareImprovement, Y.B. MED. INFORMATICS, 2000, at 66 (noting that it requires seventeen years forresearch evidence to result in clinical changes and ten years before treatment recommendationsfor new therapies appear in medical textbooks).

181. John A. Sbarbaro, Can We Influence Prescribing Patterns?, 33 CLINICAL INFECTIOUS

DISEASES S240, S241 (2001).182. For example, in 2007, after an influential medical journal published research showing

that the then-common procedure of heart surgery using stents to unclog blocked arteries wasno more effective than drugs alone, although the stent procedure briefly declined, it began toincrease again. Keith J. Winstein, A Simple Health-Care Fix Fizzles Out, WALL ST. J. (Feb.11, 2010, 1:26 PM), https://www.wsj.com/articles/SB 10001424052748703652104574652401818092212?ns=prod/accounts-wsj [https://perma.cc/MZ49-VLX2]. Of course, the doctorresistance in this case may be linked to an issue not present with drug prescription-a higherlevel of reimbursement for a procedure rather than drugs alone. However, doctors seem togenerally resist changing practices long after empirical data shows the need to do so. AaronE. Carroll, It's Hard for Doctors to Unlearn Things. That's Costly for All of Us, N.Y. TIMES:UPSHOT (Sept. 10, 2018), https://www.nytimes.com/2018/09/10/upshot/its-hard-for-doctors-to-unleam-things-thats-costly-for-all-of-us.html [https://perma.cc/VZE2-8F3K] (discussinga study indicating that doctors continue to use procedures even after they have been provedineffective).

183. Jette V. Le, Line B. Pedersen, Helle Riisgaard, Jesper Lykkegaard, Jorgen Nexoe,Jeannette Lemmergaard & Jens Sondergaard, Variation in General Practitioners'Information-Seeking Behaviour A Cross-Sectional Study on the Influence of Gender, Ageand Practice Form, 34 SCANDINAVIAN J. PRIMARY HEALTH CARE 327, 328 (2016) (findingfemale and younger physicians more likely to know about and adhere to guidelines).

[Vol. 94:773


prescription, on balance, holds more promise of doing good for thepatient than harm. 184

This may sound intuitive, but it is an intuition that may not be actually supported, asthe next Section shows. Of course, other legal academics have recognized thatdoctors are not so sophisticated, although without recognizing the existence of anoverarching schema.185

2. Debunking the Sophisticated Doctor Schema

This Section provides evidence to contradict the sophisticated doctor schema heldby judges, academics, and even doctors themselves. This Section begins by showingthat a substantial number of doctors are actually not sophisticated regarding theirknowledge of approved uses of drugs-even though they consider what uses areapproved to be important. Not only does this support the existence of a misplacedsophisticated doctor schema, but it has implications for doctors' vulnerability tomarketing they may assume is for on-label, as opposed to off-label use. Then, thisSection builds upon the fact that doctors are not sophisticated about what is anapproved use of a drug to show why doctors are far more vulnerable to marketinginfluences than they realize.

In evaluating whether doctors are able to cull out valid information from self-interested promotion about off-label uses that have not been reviewed by the FDAfor accuracy (which the sophisticated doctor schema assumes is possible), a helpfulstarting point is the extent to which doctors are knowledgeable about approved uses.After all, if a doctor cannot readily distinguish which uses are approved, it will bemore difficult for a doctor to identify when marketing focuses on unapproved usesand to be more suspect of such information. Unfortunately, a substantial number ofdoctors are confused about approved uses, even though they consider this to be animportant factor regarding what drugs to prescribe. A 2009 study found that asignificant number of doctors falsely believed some drugs uses were FDA approvedin situations that the FDA had warned against. 186 In other words, a substantial number

184. Smolla, supra note 5, at 87; see also Bagley et al., supra note 7, at 386 (arguing thatdoctors are sophisticated consumers as noted in Sorrell and thus arguably there can be no harmto any "scientific" information provided by industry representatives); Thea Cohen, The FirstAmendment and the Regulation of Pharmaceutical Marketing: Challenges to theConstitutionality of the FDA's Interpretation of the Food, Drug, and Cosmetics Act, 49 AM.CRIM. L. REV. 1945, 1967 (2012) (noting that there is trust in prescribers to make "life or deathdecisions" and be able to "evaluate health-related information, whatever its source").

185. E.g., Stephanie M. Greene, Debate, FDA Prohibitions on Off-Label Marketing DoNot Violate Drug Manufacturers' First Amendment Rights, 162 U. PA. L. REV. ONLINE 239,243 (2014) (stating that courts have mistakenly assumed that doctors are able to discernmisleading from nonmisleading information, citing the Sorrell court's reference to doctors as"sophisticated" in particular, even though that is contrary to literature showing doctors learnabout products primarily from the industry); see also Jonathan J. Darrow, PharmaceuticalGatekeepers, 47 IND. L. REv. 363, 368-69 (2014) (noting that doctors rely on companyinformation and often assume new drugs are better).

186. Donna T. Chen, Matthew K. Wynia, Rachael M. Moloney & G. Caleb Alexander,U.S. Physician Knowledge of the FDA-Approved Indications and Evidence Base for

2019]

INDIANA LAW JOURNAL

of doctors were fundamentally confused on what was FDA approved, even whenthey considered this to be relevant and important information. Similar findings havebeen reported in studies in other countries. 187

Troublingly, doctors have been found to conflate warnings against a use with theassumption that the use has actually been approved. For example, in the same 2009study, one-third of doctors thought that a drug was approved for chronic anxietywhen in fact the FDA had warned against this use, which could reflect conflation ofoff-label marketing of the drug for such use with a "black box" warning on thelabel,188 although this particular study did not focus on that issue. 189 Similarly, nearlyone in five doctors prescribed a drug for what they thought was an FDA-approveduse, when in fact at the time of the survey, that use had a black box warning forincreased risk of death.1 90 This experimental finding about black box warningconfusion is not an isolated incident; there are other situations where black boxwarnings do not appear to limit use, and sometimes use seems to increase-evenafter a manufacturer sends a letter directly to doctors. 191 Doctors could be unaware

Commonly Prescribed Drugs: Results of a National Survey, 18 PHARMACOEPIDEMIOLOGY &DRUG SAFETY 1094, 1098 (2009) (finding that nineteen percent of doctors prescribed Seroquelfor dementia with agitation believing it was FDA approved for such use when in fact the FDArequired a black box warning for use in elderly patients, and over seventy percent of doctorsprescribed lorazepam to treat chronoic anxiety assuming it was approved for this use when infact the FDA warned against such use).

187. E.g., Ghinea et al., supra note 179, at 576 (stating that Australian doctors are oftenunaware they are prescribing off label or even what that means); Constans C. A. H. H. V. M.Verhagen, Anne G. H. Niezink, Yvonne Y. Engels, Yechiel Y. A. Hekster, Joan J. Doornebal& Kris C. P. Vissers, Off-Label Use of Drugs in Pain Medicine and Palliative Care. AnAlgorithm for the Assessment ofIts Safe and Legal Prescription, 8 PAIN PRAC. 157, 158 (2008)(finding that staff members in the Netherlands were not always aware they were prescribingpain medication off label).

188. Chen et al., supra note 186, at 1098. This is the most severe warning that appears onapproved drugs and appears in a box on the label to warn of serious health risks, includinghealth-threatening risks. 21 C.F.R. § 201.57(c)(1) (2018) (noting a boxed warning may beappropriate for serious warnings and "particularly those that may lead to death or seriousinjury" and is usually based on clinical data); see also Judith E. Beach, Gerald A. Faich, F.Gail Bormel & Frank J. Sasinowski, Black Box Warnings in Prescription Drug Labeling:Results of a Survey of 206 Drugs, 53 FOOD & DRUG L.J. 403, 405-08 (1998) (stating thatalthough the FDA has not clearly indicated bases for black box warnings, empirical dataconcerning situations in which black box warnings are provided fall into six categories,primarily focused on high-risk patients and are generally based on clinical studies submittedas part of new drug applications, but possibly from postmarket reporting or epidemiologicalsurveys); Shirley Murphy & Rosemary Roberts, "Black Box" 101: How the Food and DrugAdministration Evaluates, Communicates, and Manages Drug Benefit Risk, 117 J. ALLERGY

& CLINICAL IMMUNOLOGY 34 (2006).189. Chen et al., supra note 186, at 1098.190. Id.191. E.g., Walter Smalley, Deborah Shatin, Diane K. Wysowski, Jerry Gurwitz, Susan E.

Andrade, Michael Goodman, K. Arnold Chan, Richard Platt, Stephanie D. Schech & WayneA. Ray, Contraindicated Use of Cisapride: Impact of Food and Drug AdministrationRegulatory Action, 284 J. AM. MED. ASS'N 3036, 3036-37 (2000). The public health problemonly terminated after the manufacturer voluntarily withdrew the product from market on the

[Vol. 94:773


of black box warnings,192 but other studies suggest that doctors are aware of suchwarnings and simply do not follow them. 193 So, even rigorous FDA efforts to warndoctors about drug risks may have less impact than commercial marketing.194

The sophisticated doctor schema is also well illustrated by a key 1982 study thatreveals doctors are much more influenced by commercial rather than scientificsources.195 The study chose two drugs for which there was substantial scientificliterature showing that the drugs were not effective, 196 or at least no better than over-the-counter options, and interviewed a random sample of primary care physicians.1 97

Given this, questions about the perceived value of these drugs that correlate withcommercial promotion could appropriately link actual impact of commercial forces.A substantial majority (sixty-two percent) stated that academic sources were "veryimportant" in shaping their decisions and a majority similarly claimed commercialsources to be of "minimal importance."'19 However, responses regardingscientifically unsupported statements touted by the industry indicate morecommercial influence than doctors reported. For example, nearly half of doctorsthought one drug, propoxyphene, was more potent than aspirin, contrary to scientific

eve of an FDA review of the drug's benefit versus risks, as well as after eighty reported deaths.Gardiner Harris & Eric Koli, Lucrative Drug, Danger Signals and the F.D.A., N.Y. TIMES(June 10, 2005), http://www.nytimes.com/2005/06/10/business/lucrative-drug-danger-signals-and-the-fda.html [https://perma.cc/7Z7E-ZZ2Z]; Sheryl Gay Stolberg, HeartburnDrug Linked to Deaths to Be Withdrawn, N.Y. TWEs (Mar. 24, 2000), http://www.nytimes.com/2000/03/24/us/heartbum-drug-linked-to -deaths-to -be -withdrawn.html [https://perma.cc/J5FD-DVRC].

192. Larkin et al., Restrictions, supra note 64, at 1021 (suggesting a lack of awareness ofFDA-approved indications for psychiatric drugs, as well as black box warnings as possibleexplanations for significant increase in off-label prescribing of some drugs after detailing waslimited).

193. See Supriya K. Bhatia, Amy J. Rezac, Benedetto Vitiello, Michael A. Sitorius, BruceA. Buehler & Christopher J. Kratochvil, Antidepressant Prescribing Practices for theTreatment of Children andAdolescents, 18 J. CHILD & ADOLESCENT PSYCHOPHARMACOLOGY

70, 73-75 (2008) (finding that nearly ninety-seven percent of surveyed psychiatrists wereaware of black box warnings relating to antidepressants for children, but few adhered to theFDA guidelines).

194. The persistence of the sophisticated doctor schema may be reinforced by othercognitive biases. For example, the overconfidence bias likely makes doctors assume they arenot vulnerable to marketing. In addition, the availability bias may make doctors rely more oninformation that is most repeated (i.e., commercial marketing).

195. Avorn et al., supra note 178, at 7.196. One drug was a vasodilator widely promoted for senile dementia contrary to clinical

literature showing lack of utility, whereas the other drug, propoxyphene, was promoted forsevere pain even though more than eight studies showed no advantage over a placebo. Id. at4-5.

197. Although the sample was randomly chosen among doctors in Boston who listedinternal medicine or general medicine as their area of practice, two-thirds were solopractitioners, who in other studies seem to be more influenced by commercial marketing sincethey do not have readily accessible colleagues. Id. at 5.

198. Drug advertising was considered of minimal importance to sixty-eight percent anddrug representative visits were of minimal importance to fifty-four percent. Id. Less than thirtypercent considered each to be of moderate importance. Id.

2019]

INDIANA LAW JOURNAL

data, yet consistent with commercial promotion.1 99 Similarly, almost half thoughtthat vasodilators were effective for senile dementia, which is consistent withadvertising, but inconsistent with actual scientific literature.2"' Accordingly, thisstudy indicates that doctors are substantially impacted by commercial sources, evenif the sources actually contradict what scientific studies report, contrary to doctors'claims that they value scientific studies more.2" 1 The study did not look at actualprescription data, but national prescribing data on these drugs are consistent with thestudy (i.e., the drugs are overprescribed relative to their actual utility).2" 2

A follow-up study in 1989 by one of the same authors as the 1982 study similarlydebunks the sophisticated doctor schema. The 1989 study involved the same drugsas the original study, as well as one additional drug; this second study yielded similardata.20 3 Although doctors were not specifically asked about reliance on commercialsources, more than a quarter of statements made by doctors in interviews contradictedthe scientific literature, but mirrored commercial promotions.211 Moreover, whenconfronted about the discrepancy, some doctors rejected the scientific data; onedoctor claimed clinical experience was more relevant than academic studies andothers asserted that their own "studies" prove the opposite.25 In other words,although doctors assume that they are highly sophisticated in discerning accurateinformation from promotions, not only are they vulnerable to these promotions, butthey also do not realize it.

A related problem is that, contrary to court assumptions that doctors aresophisticated enough to evaluate peer review articles presented by self-interestedcompanies, the reality is likely different. First, although doctors are trained in

199. Id.200. Id.; see also Schwartz et al., supra note 179, at 579 (finding that twenty-six percent

of doctors claimed therapeutic advantages of vasodilators on senile dementia, as well as thebenefit of propoxyphene, despite a lack of scientific support).

201. One small caveat is that although doctors generally dismissed patient preferences asnot relevant to their prescriptions, answers to some questions suggest that they wereinfluenced. See Avorn et al., supra note 178, at 6 (stating that over eighty percent of doctorsstated that they "often" prescribe a drug for pain because patients were not satisfied with over-the-counter drugs); see also J. Gambrill & C. Bridges-Webb, Use of Sources of Therapeuticand Prescribing Information by General Practitioners, 9 AUSTL. FAM. PHYSICIAN 482, 483(1980) (noting that doctors reported journals as the most useful source for prescribing, withsales representatives listed second). But see Donald P. Connelly, Eugene C. Rich, Shawn P.Curley & John T. Kelly, Knowledge Resource Preferences of Family Physicians, 30 J. FAN.PRAC. 352, 356-59 (1990) (finding that family physicians reported scientific articles to be notcredible or accessible, albeit based on self-reported data).

202. See Avorn et al., supra note 178, at 7.203. Schwartz et al., supra note 179, at 578. The study focused on possible interventions

-one involving a mailing of a visually appealing "un-advertisement" including scientific andeconomic costs and benefits to prescribing the scientifically unsupported drugs, whereas theother involved the same mailing as well as a visit by a pharmacist to educate doctors about thedrugs, suggest practical alternatives, and answer questions. The mailing group did not reduceprescriptions significantly, whereas the group that was visited in-person reduced prescriptionof the studied drugs by fourteen percent. Id.

204. See id. at 579.205. Id.

[Vol. 94:773


science, as busy individuals, they often do not read more than abstracts, which havebeen found to be inaccurate.21

6 In addition, not only is there a general publicationbias on positive outcomes, but companies will selectively present only favorablearticles to doctors,2 7 which sometimes have been "ghostwritten" by the industry.218

Although doctors should know that they are only presented with selectiveinformation, they may not actually be capable of processing that biased information.Notably, one study found that when doctors were presented with a study that reportedthe positive results of a drug based on only one of the five trials conducted, they weremore likely to prescribe the drug, even without being given the other data, let alonean opportunity to evaluate its veracity.219 In addition, once doctors have beeninfluenced by the industry, it may be difficult to change their behavior. For example,after many articles touted off-label use of a coagulant, more than ninety-sevenpercent of prescriptions were made for such uses-even after a systematic reviewconcluded that off-label use was unwarranted.21

1

Although doctors realize that the goal of sales representatives is to promote sales,they do not seem as sophisticated as courts and some scholars assume in evaluatingsuch information. They seem unaware of what information is reliable. For example,some doctors improperly assume that detailing is more reliable than printedadvertising.2" However, in reality, advertisements are regulated by the FDA whereassales representatives are generally not regulated, and they have been found to providefalse or misleading statements in detailing.21 2 In addition, although studies have

206. See Sanjay Saint, Dimitri A. Christakis, Somnath Saha, Joann G. Elmore, Deborah E.Welsh, Paul Baker & Thomas D. Koepsell, Journal Reading Habits of Internists, 15 J. GEN.INTERNAL MED. 881, 883 (2000). Another, but slightly different, problem is that abstracts donot disclose potential conflicts of interest, which could put a doctor on notice that a study wasfunded by a self-interested company. See Cdline Buffel du Vaure, Isabelle Boutron, ElodiePerrodeau & Phillipe Ravaud, Reporting Funding Source or Conflict of Interest in Abstractsof Randomized Controlled Trials, No Evidence of a Large Impact on General Practitioners'Confidence in Conclusions, a Three-Arm Randomized Controlled Trial, BIOMED CENT. MED.,

2014, at 2.207. E.g., Wash. Legal Found. v. Friedman, 13 F. Supp. 2d 51, 65 (D.D.C. 1998).208. Essentially, the industry hires freelance writers to provide a draft to a sham academic

author to publish, and journals are generally not aware of the problem since the author will notdisclose any conflict of interest. See Marcia Angell, Industry-Sponsored Clinical Research: ABroken System, 300 J. AM. MED. ASS'N 1069 (2008); Jeffrey R. Lacasse & Jonathan Leo,Ghostwriting at Elite Academic Medical Centers in the United States, PLOS MED., Feb. 2010,at 1; Joseph S. Wislar, Annette Flanagin, Phil B. Fontanarosa & Catherine D. DeAngelis,Honorary and Ghost Authorship in High Impact Biomedical Journals. A Cross SectionalSurvey, 343 BRIT. MED. J. 1192 (2011). Natasha Singer, Medical Papers by GhoswritersPushed Therapy, N.Y. TmEs (Aug. 4, 2009), https://www.nytimes.com/2009/08/05/health/research/05ghost.html [https ://perma.cc/U3WS-9QDQ].

209. Gabriel K. Silverman, George F. Loewenstein, Britta L. Anderson, Peter A. Ubel,Stanley Zinberg & Jay Schulkin, Failure to Discount for Conflict of Interest When EvaluatingMedical Literature: A Randomised Trial of Physicians, 36 J. MED. ETHICS 265, 267 (2010).

210. N. Ghinea, W. Lipworth, I. Kerridge & R. Day, No Evidence or No Alternative?Taking Responsibilityfor Off-Label Prescribing, 42 INTERNAL MED. J. 247, 248 (2012).

211. Ziegler etal., supra note 171, at 1297-98.212. 21 C.F.R. § 202.1 (2017); Beth Synder Bulik, Pharma Sales Rep Regulations

2019]

INDIANA LAW JOURNAL

shown that detailing tends to exclude negative information,113 doctors sometimeserroneously believe that negative effects are disclosed by detailers.214 The mostextreme example of lack of physician sophistication in evaluating detailinginformation is from a study that found that doctors are often unable to discern falsestatements from drug representatives.21 The representatives made statements thatwere contradicted by readily available sources, including company literature, as wellas the Physician 's Desk Reference, despite the fact that the study involved the uniquesituation where drug representatives knew that their statements were actually beingrecorded.216 Improbable claims also went unchallenged.2 17 Of particular concern, twofalse statements were not only undetected, but would result in dangerous outcomesif accepted at face value. For example, one statement claimed that there was no needto monitor for hematologic side effects, contrary to the package insert with a boxedwarning.

218

Even doctors who are skeptical about whether industry information is accuratemay nonetheless rely on inaccurate information due to the cognitive bias ofaccessibility. For example, since company detailing is more readily available andeasier to process than scientific articles, studies showing that doctors considerdetailing more "valuable" than scientific articles, even if less credible, should not besurprising. Along similar lines, advertising may have more of an impact than peerreviewed articles on prescriptions because it is similarly more accessible.219 Thisseems to be consistent with the 1982 study that indicates doctors were moreinfluenced by commercial sources than scientific sources with respect toscientifically discredited drugs. In particular, it could be that doctors believe that theyshould be influenced more by academic than commercial sources, but in reality,commercial sources are more accessible and thus more influential.

C. The More Information Schema

This Section explains the schema in commercial speech cases that favors moreinformation and why several long-standing principles in the law are not supported

Proposal Stalls in Philadelphia, but the Battle's Not over Yet, FIERCEPHARMA (Jan. 7, 2019,9:07 AM), https://www.fiercepharma.com/narketing/sales-rep-regulations-proposal-stalls-philly-but-city-vs-pharma-not-over-yet [https://perma.cc/5HAH-AS5A] (noting a failure topass legislation in Philadelphia that would require regulation of sales represenatives, and onlyChicago and the state of Nevada have passed this type of legislation thus far); see also supranote 50 (discussing misrepresentations that have contributed to the opiod epidemic); infra note283 (discussing misrepresentations made by sales representative in the Caronia case).

213. E.g., Sales Representatives: A Damning Report by Prescrire Reps MonitoringNetwork, 8 PRESCRIRE INT'L 86 (1999).

214. See Strang et al., supra note 170, at 476.215. Ziegler et al., supra note 171, at 1297.216. Id. at 1296, 1298.217. Id. at 1298.218. Id. at 1297.219. See Pierre Azoulay, Do Pharmaceutical Sales Respond to Scientific Evidence?, 11 J.

ECON. & MGMT. STRATEGY 551, 586 (2002) (finding that, in a study of H2 blockers, althoughscientific studies had some impact, marketing and especially detailing had a pronouncedeffect).

[Vol. 94:773


by empirical data. In particular, it shows how courts have relied on this schema tofavor and expand commercial speech in three ways. First, courts require a highburden of proof for allegedly misleading information. Second, courts treat potentiallymisleading information as legally interchangeable with true information. Third,courts presume disclaimers will alleviate confusion or deception. However, thisSection explains that empirical studies do not support these current commercialspeech hallmarks.

1. What Is the More Information Schema?

The more information schema is deeply embedded in commercial speech case lawthat generally favors permitting commercial speech. As discussed earlier,commercial speech case law prefers permitting information over suppression since itis presumed to provide valuable information to the audience that will promoteinformed decisions. The Washington Legal Foundation district court noted that "[i]fthere is one fixed principle in the commercial speech arena, it is that 'a State'spaternalistic assumption that the public will use truthful, nonmisleading commercialinformation unwisely cannot justify a decision to suppress it,'" 22

' and any suchrestriction is "practically an engraved invitation to have the restriction struck. 221

This preference towards more information has been consistent from the first SupremeCourt case to clearly hold commercial speech entitled to First Amendmentprotection. In that first case, the Court stated that "people will perceive their ownbest interests if only they are well enough informed, and that the best means to thatend is to open the channels of communication rather than to close them, '222 whichmany subsequent courts have cited.223

The more information schema also supports the current high burden of proof onestablishing that commercial speech is inherently misleading to permit it to beabsolutely barred without consideration of any other factors. For example, inWashington Legal Foundation, the court rejected the suggestion that scientific claimsare presumptively misleading unless approved by the FDA, which seems consistentwith the more information schema.224 As will be explained, scientific claims madeeven in peer review journals are likely to be misleading and potentially even false.Along similar lines, one court rejected the FDA's assertion that commercial claimsregarding supplements are inherently confusing when not supported by substantialscientific evidence; the FDA's contention was characterized as "almost frivolous,"

220. Wash. Legal Found. v. Friedman, 13 F. Supp. 2d 51, 69-70 (D.D.C. 1998) (quoting44 Liquormart, Inc. v. Rhode Island, 517 U.S. 484, 497 (1996)).

221. Id. at70.222. Va. State Bd. of Pharmacy v. Va. Citizens Consumer Council, 425 U.S. 748, 770

(1976) (emphasis added); see also id. at 765 (stating that the goal of commercial speech is toensure that decisions are "intelligent and well informed" and that promoting "the free flow ofcommercial information is indispensable").

223. E.g., Liquormart, 517 U.S. at 497; Bolger v. Youngs Drug Prods. Corp., 463 U.S. 60,79 (1983); Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm'n of N.Y., 447 U.S. 557, 562(1980); Am. Acad. of Implant Dentistry v. Parker, 152 F. Supp. 3d 641, 651 (W.D. Tex. 2016);Whitaker v. Thompson, 248 F. Supp. 2d 1, 8 (D.D.C. 2002).

224. Wash. Legal Found., 13 F. Supp. 2d at 67.

2019]

INDIANA LAW JOURNAL

and the court analogized the FDA claim to suggesting consumers would behypnotized to buy a product.22

In addition, the more information schema supports the current law that treatspotentially misleading information identically to truthful information that is clearlynot misleading. Notably, although courts consistently recognize harms from actuallymisleading information, it seems questionable to assume that potentially misleadinginformation is not harmful. Admittedly, the difference in treatment is largely areflection of the fact that some information can be entirely barred due to harms, evenwithout full consideration of all Central Hudson factors. However, treatingpotentially misleading information as legally equivalent to information that iscompletely truthful and not misleading seems to rely on the assumption that moreinformation-even if potentially misleading-is preferable. The more informationschema assumes that additional information from another source will correct anyinformation flaws. However, as noted earlier, that is not likely in the context ofmarketing of pharmaceutical drugs for which the marketplace is one sided.

The more information schema also explains court views on disclaimers andclarifications in the context of commercial speech decisions. In particular, courtsoften reject government assertions that commercial speech will be potentiallymisleading based on the assumption that a less restrictive-and presumed effective-alternative would be a disclaimer or warning concerning information that isconsidered misleading.226 For example, in Washington Legal Foundation, eventhough the court noted that a company would selectively provide only favorable peerreview publications to doctors with potential to mislead,2 7 the court dismissedconcerns that this could be potentially misleading for physicians and stated that thedocuments could have "conspicuous notifications" that the use being discussed wasnot approved.228 In one extreme situation, a court assumed a disclaimer is notconfusing unless there is empirical evidence of such a fact.22 9 However, as the nextSection will establish, this presumption concerning disclaimers is contrary toempirical evidence.

2. Evidence Debunking "More Information Is Better" Schema

This Section provides empirical evidence to contradict the seemingly logicalpresumption that more information provides more informed results. Consistent withthe earlier discussion of cognitive limits, this Section explains that there are seriouscognitive limits to processing disclaimers, which is very important to the popular

225. Pearson v. Shalala, 164 F.3d 650, 655 (D.C. Cir. 1999); see also Int'l Dairy FoodsAss'nv. Boggs, 622 F.3d 628, 636-37, 639 (6th Cir. 2010) (relying on amici information anddiscounting the FDA view that a statement that milk from cows not treated with hormonemight improperly suggest milk is safer and asserting that any possible problem withmisleading claims is alleviated by a disclaimer).

226. E.g., United States v. Caronia, 703 F.3d 149, 168 (2d Cir. 2012).227. Wash. Legal Found., 13 F. Supp. 2d at 65 (stating this in the context of concluding

that distribution of scientific articles constitutes commercial speech, rather than scientific andacademic speech that is deserving of more First Amendment protection).

228. Id. at 68.229. Whitaker v. Thompson, 248 F. Supp. 2d 1, 13 (D.D.C. 2002).

[Vol. 94:773


First Amendment solution of providing disclaimers as a less speech restrictive meansthat allegedly will protect against misleading information.3 Although there are noknown studies of disclaimers regarding off-label drug use, there are studies in tworelated contexts that indicate disclaimers do not result in more informed decisions:disclaimers to health claims for dietary supplements as well as disclaimers in articlesregarding industry funding.

Although disclaimers on dietary supplements may initially seem less analogousto off-label promotion of drugs to doctors since supplement disclaimers focus onconsumers, supplement disclaimers importantly refer to the lack of FDA reviewregarding health benefits, which is analogous to off-label uses of drugs that have notbeen approved by the FDA.23 1 In particular, dietary supplement companies can makeclaims lacking significant scientific agreement if they include a disclaimer that thehealth claim(s) are not reviewed by the FDA and that the supplement is not intendedto treat any condition or disease.23 2 However, a recent systematic review of seventeenpublished articles to empirically test a court suggestion that disclaimers couldeffectively warn doctors regarding unapproved uses of drugs concluded thatconsumers, including those that use supplements with disclaimers, are generallyunaware of their existence, or that they did not attach any weight to the informationin the disclaimer, as will be discussed.2 33

230. Courts seem to intuit this in another context; in particular, in products liability casesinvolving failure to warn, courts have suggested that more warnings would not be better dueto information overload. Hood v. Ryobi Am. Corp., 181 F.3d 608, 611 (4th Cir. 1999) (notingthat too many warnings make a label too long to read and too technical to understand); JamesA. Henderson, Jr. & Aaron D. Twerski, Doctrinal Collapse in Products Liability: The EmptyShell of Failure to Warn, 65 N.Y.U. L. REv. 265, 297-98 (1990) (noting that judges and jurieserroneously assume warnings are costless in terms of harm). In addition, recent scholarshipsuggests that although many areas of the law suggest more information is better, consumersare not well poised to understand compelled disclosures. E.g., Omri Ben-Shahar & Carl E.Schneider, The Failure ofMandatedDisclosure, 159 U. PENN. L. REv. 647, 718 (2011); KestenC. Green & J. Scott Armstrong, Evidence on the Effects of Mandatory Disclaimers inAdvertising, 31 J. PUB. POL'Y & MARKETING 293, 302 (2012).

23 1. As a result of changes in the law that permit dietary supplements to be sold withoutthe rigorous review of drugs, as well as First Amendment challenges, the FDA currentlypermits claims that lack significant scientific agreement to be made with certain disclaimers.Guidance for Industry: Interim Procedures for Qualified Health Claims in the Labeling ofConventional Human Food and Human Dietary Supplements, FOOD & DRUG ADMIN. (July2003), https ://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/LabelingNutrition/ucm53832.htm [https://perma.cc/VTA4-HGAK]. This isrelatively recent, given changes in the regulatory framework in the mid-1990s, followed byFirst Amendment challenges. See Dietary Supplement Health and Education Act, Pub. L. No.103-417, 108 Stat. 4325 (1994); Whitaker, 248 F. Supp. 2d at 2; Pearsonv. Shalala, 164 F.3d650, 655-60 (D.C. Cir. 1999).

232. 21 C.F.R. § 101.93(c) (2018) ("This statement has not been evaluated by the Foodand Drug Administration. This product is not intended to diagnose, treat, cure, or prevent anydisease.").

233. E.g., Kesselheim et al., Mandatory Disclaimers, supra note 96, at 444 (evaluatingsuggestion noted in United States v. Caronia, 703 F.3d 149, 179-80 (2d Cir. 2012)).

2019]

INDIANA LAW JOURNAL

Multiple studies established that disclaimers on supplements had no effect onconsumer belief regarding whether the product would have the claimed effect, or onthe belief that the FDA had evaluated the claims.2 34 The largest qualitative study todate involving 3500 adults indicated that about half of all surveyed adults believedthat over-the-counter herbal products and weight-loss supplements are approved bya government agency, even though these products have disclaimers that clearly statethey have not been reviewed by the FDA for effectiveness.2 35 In addition, studiesfound that even for supplement-using consumers aware of disclaimers, some stillbelieved that the FDA had tested the supplements or otherwise verified the accuracyof manufacturer claims. 2 36 Although supplement users may have cognitive biases thatmotivate them to believe in the safety of products they are using, the fact that evennonusers assume the FDA verifies claims suggests that such bias does not completelyexplain failure to understand FDA disclaimers.2 37 Rather, it may be that in an

234. E.g., Tonya Dodge & Annette Kaufman, What Makes Consumers Think DietarySupplements Are Safe and Effective? The Role of Disclaimers and FDA Approval, 26 HEALTH

PSYCHOL. 513, 516 (2007) (finding that FDA-required disclaimer resulted in only slightreduction in effectiveness belief with disclaimer, but no impact on beliefs about risks); MaryMeer, Scottie Misner & Ralph Meer, Labeling of Dietary Supplements: Consumer Awarenessand Industry Compliance, 4 J. NUTRACEUTICALS FUNCTIONAL & MED. FOODS 29, 35 (2004)(finding that almost thirty percent of study participants were unaware of the existence of adisclaimer); Carla K. Miller & Teri Russell, Knowledge of Dietary Supplement LabelInformation Among Female Supplement Users, 52 PATIENT EDUC. & COUNSELING 291, 294-96 (2004) (in a study of dietary supplement users, two-thirds thought that the product had beenapproved by the FDA); Karen Russo France & Paula Fitzgerald Bone, Policy Makers'Paradigms and Evidence from Consumer Interpretations of Dietary Supplement Labels, 39 J.CONSUMER AFF. 27, 47 (2005) (finding that a disclaimer appeared to have no effect on beliefthat the product would have the claimed effect or that the FDA had reviewed the claimedeffectiveness).

235. Janine L. Pillitteri, Saul Shiffman, Jeffrey M. Rohay, Andrea M. Harkins, Steven L.Burton & Thomas A. Wadden, Use of Dietary Supplements for Weight Loss in the UnitedStates: Results of a National Survey, 16 OBESITY 790, 791, 793 (2008) (finding that 48.8% ofself-identified supplement users and 45% of nonusers agreed with statement that over-the-counter herbal products and weight-loss supplements must be approved for efficacy, oreffectiveness by a government agency).

236. E.g., Marlys J. Mason & Debra L. Scammon, Unintended Consequences of HealthSupplement Information Regulations. The Importance ofRecognizing Consumer Motivations,45 J. CONSUMER AFF. 201, 211 (2011); see also Miller & Russell, supra note 234, at 294(noting that when presented with an actual disclaimer, more than thirty percent of participantsfailed to understand the FDA had not specifically approved or regulated the product).

237. In particular, supplement users are likely to have an over optimism bias that leadsthem to believe products they use are safer than they are. See Robert J. Blendon, Catherine M.DesRoches, John M. Benson, Mollyann Brodie & Drew E. Altman, Americans' Views on Useand Regulation of Dietary Supplements, 161 ARCHIVE INTERNAL MED. 805, 808 (2001). Inaddition, to the extent that they have a general belief that supplements are safe, that assumptionwill remain in the face of contrary evidence pursuant to confirmation bias, which seems to bereflected in studies. See id. at 807 (finding sixty-seven percent of supplement users indicatedthat if the FDA said a supplement was ineffective, that would not impact their decision tocontinue to take the supplement); Mason & Scammon, supra note 236, at 210, 214-15 (findingthat supplement users who understood the FDA does not regulate supplements focused on

[Vol. 94:773


information-rich world already saturated with legal disclaimers, consumers areinured to them.2138 In addition, although disclaimers in real life could beovershadowed by other marketing messages, including more prominent images, suchthat there is incomplete processing3 9 even in controlled experimental settingswithout competing information or the presence of brand loyalty to a specificsupplement, individuals still fail to understand the disclaimers.

Admittedly, the studies about disclaimers on supplements and lay consumers maynot seem directly applicable to the allegedly more sophisticated doctors. However,studies consistently show that all individuals, including highly educated ones likedoctors, are subject to the same cognitive biases. In addition, there are specificstudies indicating physician behavior mirrors consumer behavior in marketingcontexts, such that other researchers have suggested that supplement studiesregarding consumers are analogous given the lack of any studies regardingphysicians in particular.24 In addition, there are studies on doctors in particular, andthe impact of conflict of interest disclosures, as discussed below, that suggest thereare serious limitations to the use of disclaimers to adequately inform doctors.

Accordingly, if companies present their alleged facts to doctors who arevulnerable to accepting such facts given the lack of competing information, anydisclaimer to doctors about the lack of FDA review may similarly fall on deaf ears.This problem is compounded by the cognitive bias of anchoring in that the initialinformation from companies will be given more weight, or at least weight that isdisproportional to the actual data. In addition, to the extent that companies tend toflood doctors with promotional materials, this information will seem more availableand thus more relevant and accurate according to cognitive bias studies. However,all individuals, including experts, are subject to cognitive biases, and doctors arevulnerable to overconfidence bias that would likely make them even less likely torealize the impact of these heuristics. In addition, although the threat of a malpracticeclaim may theoretically seem a promising avenue to guard against bad decisions, thatseems inconsistent with literature about the strength of cognitive biases. Moreover,malpractice claims are inherently tied to standard of care, and if there is no clearstandard of care involving some off-label uses, malpractice threats would seem to beof little impact.

Just as disclaimers on nutritional supplements are intended to help provideinformation to ensure adequate discounting of possible bias, medical journals informdoctors about conflicts of interest to ensure that messages are appropriatelydiscounted.21 However, despite a more expert audience, there seem to be similar

prior problems with FDA-regulated products and attributed disclaimers as some legal orbureaucratic necessity).

238. E.g., Ben-Shahar & Schneider, supra note 230, at 689-90.239. See, e.g., Paul Biegler & Patrick Vargas, Ban the Sunset? Nonpropositional Content

and Regulation of Pharmaceutical Advertising, 13 AM. J. BIOETHICS 3, 8 (2013) (notingconsumers may be more influenced by positive associations suggested in advertisement thanspecific claims).

240. E.g., Kesselheim et al., Mandatory Disclaimers, supra note 96, at 440.241. Eg., Editorial Policies, NEW ENG. J. MED., https://www.nejm.org/about

-nejm/editorial-policies [https://perma.cc/HRF3-JB7H] (providing guidelines to ensurepublication is free from commercial influence); Instructions for Authors, JAMA

2019]

INDIANA LAW JOURNAL

problems processing these disclosures-even when they are actually made.242 Thedisclosure statement may be entirely ignored,4 3 or have the ironic effect ofenhancing the trustworthiness of the discloser.244 Although studies sometimesindicate doctors are more skeptical of industry-funded articles,45 perhaps because

https ://j amanetwork.com/j ournals/jama/pages/instructions-for-authors#SecDecisionsandManagementofEditorialConflictsofinterest [https://perma.cc/MR2Z-5XHW] (providingguidance on conflicts of interest); see also Phil B. Fontanarosa & Howard Bauchner, Conflictof Interest and Medical Journals, 317 J. AM. MED. ASS'N 1768, 1768469 (2017) (explainingpolicy of all JAMA-related journals to mandate disclosure of conflicts regardless of amountof funding that will be disclosed in the acknowledgements section); INTERNATIONALCOMMITTEE OF MEDICAL JOURNAL EDITORS, Recommendations for the Conduct, Reporting,Editing, and Publication of Scholarly Works in Medical Journals, INT'L COMM. MED. J.EDITORS, Dec. 2017, at 3-4, http://www.icmje.org/icmje-reconmendations.pdf[https://perma.cc/2VLA-9BWB] (providing disclosure guidelines that many medical journalsfollow). Disclosure of conflicts is also an issue with continuing medical education. E.g.,ACCREDITATION COUNCIL FOR CONTINUING MED. EDUC., STANDARDS FOR COMMERCIAL

SUPPORT: STANDARDS TO ENSURE THE INDEPENDENCE OF CME ACTIVITIES 2 (2014) (statingthat content not promoting a specific commercial interest and source of support fromcommercial interest must be disclosed to learners). However, there are some that believemerely identifying conflicts is not enough. E.g., AM. ACAD. FAMILY PHYSICIANS, CME POLICYAND PROCEDURES FOR FULL DISCLOSURE AND IDENTIFICATION AND RESOLUTION OF CONFLICTS

OF INTEREST paras. 4(B), (D), https://www.aafp.org/fpm/forms/coipolicy.pdf[https://perma.cc/3E8Q-8CE8] (requiring that conflicts not merely be identified, but resolved,which could involve assigning a different topic for the individual, or else ensuring validcontent through prior peer review or using evidence-based content); Bernard Lo & ChelseaOtt, What is the Enemy in CME, Conflicts of Interest or Bias?, 310 J. AM. MED. ASS'N 1019,1019-20 (2013) (suggesting that identification of conflicts is only a rough proxy forcommercial bias, and suggesting improvements to assess bias).

242. Sometimes articles are "written" by doctors paid by the industry, but this is notdisclosed. See Aaron S. Kesselheim, Bo Wang, David M. Studdert & Jerry Avorn, Conflict ofInterest Reporting by Authors Involved in Promotion of Off-label Drug Use: An Analysis ofJournal Disclosures, PLoS MED., 2012, at 1, 7.

243. Silverman et al., supra note 209.244. Steven D. Pearson, Ken Kleinman, Donna Rusinak & Wendy Levinson, A Trial of

Disclosing Physicians'F inancial Incentives to Patients, 166 ARCHIVES INTERNAL MED. 623,625-26 (2006).

245. E.g., Aaron S. Kesselhein, Christopher T. Robertson, Jessica A. Myers, Susannah L.Rose, Victoria Gillet, Kathryn M. Ross, Robert J. Glynn, Steven Joffe & Jerry Avorn, ARandomized Study of How Physicians Interpret Research Funding Disclosures, 367 NEw ENG.J. MED. 1119, 1122 (2012) (doctors in experimental conditions were less likely to considerindustry-funded trials as rigorous or inclined to prescribe drugs based on these studiescompared to studies purportedly from the NIH based on the abstracts of hypothetical drugs);see also Samena Chaudhry, Sara Schroter, Richard Smith & Julie Morris, Does Declarationof Competing Interests Affect Readers'Perceptions? A Randomised Trial, 325 BRIT. MED. J.1391, 1392 (2002) (subjects finding article that disclosed conflict of interest to be"significantly less interesting, important, relevant, valid, and believable" compared to articlewith no such disclosure); Sara Schroter, Julie Morris, Samena Chaudhry, Richard Smith &Helen Barratt, Does the Type of Competing Interest Statement Affect Readers'Perceptions ofthe Credibility of Research? Randomised Trial, 328 BRIT. MED. J. 743, 743 (2004) (findingthat readers of British Medical Journal considered a study to be less important, relevant, valid,

[Vol. 94:773


they are aware that such articles disproportionately yield positive results, 246 otherstudies suggest that disclosure of industry funding or other conflict of interest do notimpact how articles are perceived.247 Most notably, one study showed that althoughdoctors say they find industry-funded studies less persuasive, in an experimentalsetting regarding an abstract of a fictitious new drug, disclosure of funding from acompany had no significant impact on self-reported prescribing likelihood. 248 Indeed,almost forty percent reported that they would not consider any type of conflict ofinterest by the author in deciding what to prescribe.24 9 In other words, doctors seemto believe that they should discount for conflicts of interest, but have difficulty doingso; rather, doctors did not discount for bias at all. This is consistent with literatureshowing that people have difficulty ignoring information, such as a conflict ofinterest, in making decisions, even if they know that this information may bias theirjudgment.211 In addition, this effect could also be a function of the fact that manydoctors have relationships with pharmaceutical representatives and are motivated,albeit unconsciously, to ignore conflicts with pharmaceutical companies to preserveany potential cognitive dissonance.

There is additional data to debunk the more information schema in the context ofconflict of interest disclosures in peer review articles that are presumed to lead tomore reasoned judgments. Doctors generally only read abstracts,21 which typicallydo not include conflict of interest disclosures since these usually appear at the end of

or believable when associated with a financial interest disclosure). Of course, there are somethat suggest financial interests alone do not necessarily indicate whether a study is reliable.E.g., David B. Resnik & Kevin C. Elliott, Taking Financial Relationships into Account whenAssessing Research, 20 ACCOUNTABILITY IN RES.: POLICIES & QUALITY ASSURANCE 184, 194(2013) (suggesting several factors be used to assess whether financial relationships impactresearch credibility).

246. See supra note 60 and accompanying text.247. One study of actual abstracts from industry-funded trials indicated no statistically

significant difference in interpretation of abstract findings regarding quality of methodologyor treatment benefit. Buffel du Vaure et al., supra note 206, at 7-8.

248. See Silverman et al., supra note 209, at 267-68.249. Id. at 267.250. See, e.g., Lee Ross, Mark R. Lepper & Michael Hubbard, Perseverance in Self-

Perception and Social Perception: Biased Attributional Processes in the DebriefingParadigm, 32 J. PERSONALITY & SOC. PSYCHOL. 880 (1975); Timothy D. Wilson & NancyBrekke, Mental Contamination and Mental Correction. Unwanted Influences on Judgmentsand Evaluations, 116 PSYCHOL. BULL. 117 (1994).

251. See Saint et al., supra note 206, at 883 (finding internists reported only readingabstracts for two-thirds of articles and suggesting that this may be an underestimation); seealso Sally Hopewell, Anne Eisinga & Mike Clarke, Better Reporting of Randomized Trials inBiomedical Journal and Conference Abstracts, 34 J. INFO. SCI. 162, 162-63, 166 (2008)(noting that doctors may make treatment decisions based on reading abstracts); Robertson,supra note 60, at 368 (stating that physicians commonly choose to read only abstracts ofarticles through services such as PubMed). In addition, abstracts also may provide inaccurateconclusions. Roy M. Pitkin, Mary Ann Branagan & Leon F. Burmeister, Accuracy of Data inAbstracts of Published Research Articles, 281 J. AM. MED. ASS'N 1110, 1110-11 (1999)(noting that eighteen to sixty-eight percent of abstracts report findings inconsistent with orabsent from the article's body, even in large-circulation medical journals).

2019]

INDIANA LAW JOURNAL

an article. So, the inclusion of this additional information seems to realistically haveno chance of actually informing doctors-even if the doctor was aware that industry-supported studies tend to be correlated with positive studies.252 Moreover, even inthe unlikely situation that a doctor reads the entire article, studies suggest that theabstract could have predominant influence pursuant to cognitive science studiesregarding the effects of "anchoring"; in other words, initial information is given moreweight, even if other information follows. This would also seem to be true in theunlikely event that a doctor skipped the abstract and read the article; since the conflictof interest disclosures do not come until the end, the doctor's opinion is likely alreadyanchored to the information disclosed before such disclosures. Moreover, even ifdoctors actually read the conflict of interest disclosure and were appropriatelyskeptical of industry-funded articles, that might be of no utility if industry-fundedarticles are the only source of information.

III. IMPLICATIONS

This Part evaluates how commercial speech law currently overprotects marketingof off-label uses of drugs. Section A explains why the previously unrecognizedschemas discussed above have contributed to an expansion of commercial speechrights in this area. Then, Section B examines whether there are better FirstAmendment policies to emphasize with such promotion.

A. Retrospective View on First Amendment Expansion Informed by Cognitive Bias

This Section takes a fresh look at recent commercial speech cases regarding off-label marketing of prescription drugs. This Section first reviews key facts from recentcases. Then, it examines how schemas have impacted evaluation of facts, as well asapplication of the law. Finally, this Section explains how FDA capitulation tojudicialchallenges has resulted in the fact that schemas are currently baked into precedent.

252. There is consistent evidence that studies financially supported by self-interestedindustries are more likely to provide outcomes favorable to those industries. E.g., Justin E.Bekelman, Yan Li & Cary P. Gross, Scope and Impact of Financial Conflicts of Interest inBiomedical Research: A Systematic Review, 289 J. AM. MED. ASS'N 454, 456 (2003); JoelLexchin, Lisa A. Bero, Benjamin Djulbegovic & Otavio Clark, Pharmaceutical IndustrySponsorship andResearch Outcome and Quality: Systematic Review, 326 BRIT. MED. J. 1167,1168-69 (2003); Andreas Lundh, Joel Lexchin, Barbara Mintzes, Jeppe B. Schroll & LisaBero, Industry Sponsorship and Research Outcome, COCHRANE LIBR., 2013, at 12; SergioSismondo, Pharmaceutical Company Funding and Its Consequences: A QualitativeSystematic Review, 29 CONTEMP. CUNICAL TRIALS 109 (2008); see also Bodil Als-Nielsen,Wendong Chen, Christian Gluud & Lise L. Kjaergard, Association of Funding andConclusions in Randomized Drug Trials: A Reflection of Treatment Effect or Adverse Events?,290 J. Am. MED. ASS'N 921, 925 (2003) (noting that studies supported by for-profitorganizations are more likely to recommended experimental drug for treatment).

[Vol. 94:773


1. Relevant Case Law Expanding Commercial Speech for Off-Label Promotion

Before analyzing the impact of schemas on how courts evaluate facts and conductlegal analysis, a brief review of the core facts of cases is essential. As discussedearlier, the first commercial speech case involving off-label promotion was the 1998Washington Legal Foundation decision2 3 In that case, two FDA guidancedocuments that barred pharmaceutical companies from showing doctors scientificarticles relating to drugs were deemed unconstitutional for restricting free speech.254

The FDA did not challenge this district court decision.255 Accordingly, the nextimportant commercial speech case involving pharmaceutical companies is the 2011Supreme Court Sorrell decision2156

In 2011, the Supreme Court issued the Sorrell decision, which is related to off-label promotion of drugs but does not squarely address the extent to which the FDAcan regulate such promotion consistent with commercial speech.2157 Sorrell involveda challenge by drug companies to a Vermont law barring companies from buyingdata concerning doctor prescriptions to use it to more effectively sell drugs to doctors(i.e., for detailing by representatives).2

' A 6-3 majority found that the Vermont lawunduly hindered commercial speech and even suggested that commercial speechmight be subject to heightened judicial scrutiny if it is content or speaker based.215 9

Although the majority found that Vermont's law targeted only pharmaceuticalcompanies, and thus was arguably speaker based, it did not ultimately propose anynew standard and instead found the Vermont law failed under the traditional CentralHudson test.2 6 °

The first true First Amendment challenge to off-label marketing was in UnitedStates v. Caronia, which involved a criminal prosecution for misbranding.2 61 Thecase involved an unusual situation where the pharmaceutical sales representative wasnot only caught on tape promoting the drug for an unapproved uses, but also forsituations subject to a black box warning.262 In other words, the sales representativewas actually providing untruthful information. The FDA, however, made thestrategic decision not to challenge the falsity of the information. The Second Circuitmajority opinion in Caronia reversed the conviction against the sales representative,

253. Wash. Legal Found. v. Friedman, 13 F. Supp. 2d 51 (D.D.C. 1998).254. Id. at 54, 73-74.255. The FDA and Washington Legal Foundation (WLF) reached an agreement that

guidance documents were merely "safe harbors" from prosecution that mooted the need for anappeal of this decision as well as a related challenge to a new law permitting manufacturerdissemination of off-label use if the manufacturer had applied or intended to apply for a newuse. Krause, supra note 83, at 409.

256. Sorrellv. IMS Health Inc., 564 U.S. 552 (2011).257. See id.258. Id. at 558-61.259. Id. at 580.260. Id. at 572.261. 703 F.3d 149 (2d Cir. 2012).262. Id. at 155-57 (although the black box warning stated that drug safety and efficacy had

not been established for those under sixteen and that the drug had "very limited" experienceamong elderly patients, the sales representative was recorded saying that the drug was "a verysafe drug" to patients, including those as young as fourteen and greater than sixty-five).

2019]

INDIANA LAW JOURNAL

finding that the FDA improperly relied on off-label promotion alone as the basis ofthe misbranding claim, rather than as evidence of intent to misbrand.26 3

Subsequently, pharmaceutical manufacturer Amarin, together with four doctors,sought a preliminary injunction against the FDA in the Southern District of NewYork, where Caronia was precedential.264 In particular, Amarin wanted the court toenjoin the FDA from bringing a misbranding action265 for an unapproved use of adrug after the FDA denied approval for that use as not scientifically justified.266

Although only the FDA can approve a new use, Amarin's legal challenge waseffectively an attempt to overrule the FDA decision denying approval. Amarin,however, characterized its effort as simply an attempt to shield itself fromprosecution for making truthful statements in a format different than the FDAprefers.

267

2. Examining Key Cases Through the Lens of the Revealed Schemas

Examining recent commercial speech cases concerning off-label promotionthrough the lens of the two schemas provides new insight to complement existingcriticism of these cases. As will be discussed, these schemas impact how courtsevaluate facts, as well as the application of relevant law.

Schemas serve as an important framework through which facts are evaluated. Thisis well illustrated in the Sorrell majority opinion which found a Vermont law barringpharmaceutical companies from obtaining and using doctor prescriptions to informtheir detailing to violate protections on commercial speech.268 The majority statedthat there were "divergent views" regarding the value of detailing, as well as thevalue of prescribing brand name drugs, but nonetheless seemed more inclined toassume the information valuable and consistent with the more information schema.269

This schema likely led the majority to fail to recognize the inherently unbalancedprescription marketplace where companies have outsized resources and incentives topromote a view that cannot be matched by a state.270 The dissent, on the other hand,

263. Id. at 168-69.264. AmarinPharma, Inc. v. FDA, 119 F. Supp. 3d 196 (S.D.N.Y. 2015); see also Krause,

supra note 83, at 414-15 (noting that the case was intentionally filed in a district whereCaronia would be precedential).

265. See Amarin, 119 F. Supp. 3d at 221.266. Id. at 209 (noting desire to promote unapproved drug use for patients with merely

"persistently" high triglyceride levels in addition to the approved use for patients with veryhigh levels).

267. Id. at 223. This case was different than the prior cases not only because of itsprocedural posture, but also because the drug at issue, marketed as Vascepa, is actually apurified version of fish oil, and supplements made of unpurified fish oil may be marketedwithout the rigorous review required of drugs, so long as a company provides a disclaimer tonote the lack of a review. See id. at 237. In addition, although the FDA had a legal basis todeny approval of the proposed new use, Amarin had entered into a special agreement with theFDA before doing the clinical trial for the proposed new use that typically leads to approval.Id. at 210.

268. Sorrell v. IMS Health Inc., 564 U.S. 552 (2011).269. Id. at 578.270. See id.

[Vol. 94:773


recognized that the majority's suggestion that the state could promote contraryspeech was not realistic, citing information in the record to counter the majority'spresumptions .21 The more information schema may also underlie the majority'semphasis on the possible value of new drugs that are subject to detailing-when theissue was not about promoting all new drugs. Rather, as recognized by the dissent,the challenge was only regarding whether companies could more successfully marketnew drugs not based on a superior product, but on knowledge of doctorpreferences.

72

The schemas also play an important role in how courts evaluate regulations thatimpact commercial speech. The traditional benchmark for evaluating such speech isthe intermediate scrutiny test announced in Central Hudson. Althoughthe Sorrell majority suggested heightened scrutiny should apply to any regulationthat it is not content neutral (i.e., allegedly targeting only pharmaceuticalcompanies)273 it did not articulate, let alone apply, such a test nor have anysubsequent cases, as noted earlier. Accordingly, how schemas impact judicialevaluation of the Central Hudson factors remains an issue. Examples of the impactof schemas on judicial interpretation of the factors in key cases relating to promotionof off-label use of drugs are discussed below within the framework of the CentralHudson factors.

i. False or Inherently Misleading?

The schemas have played a role in judicial analysis of the first Central Hudsonfactor (i.e., whether the commercial speech is aimed at a lawful purpose and whetherthe speech is false or inherently misleading).

The first case to embrace First Amendment challenges to off-label use restrictions,Washington Legal Foundation, seemed influenced by these schemas in rejecting theFDA suggestion that publications from companies were false or inherentlymisleading. The court acknowledged that the FDA had provided the court with"substantial evidence" that research from companies impacted actual prescriptions

271. Id. at 593-95. As recognized by the dissent, attempts to mirror the methods of theprescription drug industry with "academic" or "counter" detailing (i.e., a presentation by anindividual to a doctor on the benefits of generic drugs or alternative treatments) have not beenfound adequate to counter the imbalanced sales tactics at issue here. Id. at 601 (Breyer, J.,dissenting).

272. See id. at 595, 596 (Breyer, J., dissenting) (noting that because, as the majority stated,state law intended to impose a "specific, content-based burden," heightened judicial scrutinyis warranted in evaluating government regulation of speech because of disagreement with themessage). Arguably, this suggestion could be a result of the convergence of the sophisticateddoctor and more information schemas. Of course, this is not to suggest that these schemas arethe only possible explanation. Alternatively, the majority could be reflecting a return to theLochner era where courts substituted their own judgment in evaluating regulations. See id. at591-93 (Breyer, J., dissenting) (cautioning that second-guessing legislative decisions wouldbe a dangerous return to the "happily bygone era" in Lochner whenjudges abused their powerto impose their own economic theories). Nonetheless, recognizing the effect of combiningboth schemas in this case may help explain the majority's suggestion for this new standard.

273. Id. at 565.

2019]

INDIANA LAW JOURNAL

and sales,2 74 and even acknowledged that manufacturers would only "disseminateinformation that presents their product in a favorable light," with the potential tomislead in the context of concluding that the information was commercial speech.2 7

However, the court did not mention these issues in analyzing whether the informationwas in fact inherently misleading, even though it acknowledged that determiningwhether something is inherently misleading depends upon possibilities for deception,which its earlier discussion seemed to suggest is possible.2 76 Even though the courtrealized that companies would likely selectively provide certain data to companies,the court asserted that publications could not be inherently deceptive when providedby one self-interested company versus another.2 77 Although the court did notexplicitly say so, it may have been relying on the sophisticated doctor schema toassume that doctors would know publications from companies were not reliable andthus perhaps merely "potentially," rather than inherently, misleading. 2 78

This lack of concern for deception may be tied to another issue the court notes asimportant in evaluating what is inherently misleading-the ability of the audience toevaluate the claims made.2 79 Although doctors cannot actually evaluate any of thescientific claims on their own without conducting their own research, which is highlyunlikely, the court's lack of concern that doctors will be misled seems to be tied tothe sophisticated doctor schema. For example, in noting that a prior court had denieda constitutional challenge to FDA regulations requiring health claims to be supportedby significant scientific agreement, the court asserted that the case at issue concerneda "professional audience," without discussion of whether that audience couldactually evaluate the scientific claims.280 In addition, the more information schemamay have influenced the court to conclude that publications from a manufacturerwere not inherently misleading because the FDA could indicate that uses had notbeen approved and require publications only from "bona fide peer-review" journalsand independent publishers, presumably based on the assumption that such sourcesare scientifically reliable.281

The more information schema may have played a particularly powerful role in theCaronia decision that overruled a criminal conviction against a sales representativefor misbranding, based in part from promoting an off-label use that could have ledto serious public health harm and was subject to a black box warning.282 This schemamay explain something that has perplexed scholars-the fact that the majorityrepeatedly emphasized that the information was "truthful," when in fact it was

274. Wash. Legal Found. v. Friedman, 13 F. Supp. 2d 51, 63 (D.D.C. 1998).275. Id. at 65.276. See id. at 66-69.277. See id.278. The court stated "the exact same journal article or textbook reprint cannot be

inherently conducive to deception and coercion when it is sent unsolicited" by a manufacturer.Id. at 67. However, this seems contrary to the court's prior acknowledgement thatmanufacturers will, in fact, selectively send only information favorable to their product.

279. See id. at 67.280. Id. at 68.281. Id.282. United States v. Caronia, 703 F.3d 149, 155-56 (2d Cir. 2012).

[Vol. 94:773


patently false.283 Of course, it is true that the FDA never alleged that the informationwas false, such that the court was legally permitted to infer the information wastruthful.2 s4 In addition, in the preliminary discussion of facts, the majority did notethat the sales representative promoted uses with populations that the FDA not onlyhad not approved, but for which there was a "black box" warning; such a warning isthe most serious warning possible for prescription labels s.2 5 However, the moreinformation schema could help explain why the court did not explicitly state in thetext to its Central Hudson analysis that it was presuming the information was notfalse or misleading because it was uncontested even though the discussion of factsacknowledged the information was not only false, but dangerous.2

16 In other words,

this case demonstrates how the more information schema not only results in treatingpotentially misleading information equivalent to truthful information, but that it mayalso result in a court presuming information is truthful contrary to actual facts. Eventhough the court was permitted as a matter of law to presume the information wastrue, the more information schema provides an important explanation for why thecourt seemed to discount the facts in its own opinion that contradict this assumption.

ii. Substantial Government Interest?

The same schemas may have contributed to the court's decision in WashingtonLegal Foundation that no substantial government interest in ensuring doctors haveaccurate and unbiased information. In particular, although the court recognized thatthere was a substantial government interest in encouraging companies to submit newdrug uses to the FDA for approval28 7 the court completely rejected the FDA'ssuggestion that there was a substantial government interest in ensuring doctors hadaccurate and unbiased information.288 This rejection seems grounded in the moreinformation schema that assumes even potentially misleading information can be"informative," as well as the sophisticated doctor schema that erroneously assumesdoctors will not be misled by marketing information.

Similarly, the schemas may have contributed to the Sorrell majority narrowlydefining the relevant government interest behind the state law intended to bar drug

283. Krause, supra note 83, at 432 (noting that the court's conclusion was "curious" sinceCaronia's statement directly contradicted the black box warning); see also Caronia, 703 F.3dat 166, 168.

284. This likely reflected the litigation strategy since a holding that does not rely on falsestatements would be broader. See Greene, supra note 5, at 683 (noting that the governmentbelieved it only needed to show that the drug was promoted for off-label use).

285. Caronia, 703 F.3d at 155-56; see also supra note 262 and accompanying text(providing details of the black box warning); supra note 188 and accompanying text(explaining black box warnings).

286. See id. Of course, the existence of an information schema does not mean that thereare not additional issues contributing to the decision, such as that the court might have beenconcerned about criminal liability for a sales person for misbranding, as opposed to acorporation or a high-level executive. E.g., id. at 174 (Livingston, J., dissenting). Nonetheless,the information schema can add insight.

287. Wash. Legal Found. v. Friedman, 13 F. Supp. 2d 51, 71-72 (D.D.C. 1998).288. Id. at 69-70.

2019]

INDIANA LAW JOURNAL

companies from using doctor prescription records. In particular, in discussing thegovernment interest, the majority characterized the main interest as medical privacy,as well as preventing harassing behavior by sales representatives.28 9 In addition, themajority relied on the sophisticated doctor schema to dismiss the suggestion that thestate law did not directly advance the goal of minimizing harassing behavior becausedoctors can and do deny meetings with detailers.29 The majority also embraced themore information schema in arguing that detailers had valuable information fordoctors-even though the information sought was only valuable to the detailers topromote sales, and not doctors.2 91 The majority discounted a state interest in limitingthe success of pharmaceutical detailing to help reduce drug costs because the statedid not vigorously pursue this during oral argument.292 However, there was asubstantial legislative record behind the policy goal of reducing detailing success toreduce costs, as duly noted by the dissent.2 93 The majority seemed to thoroughlyembrace the more information schema in characterizing detailing as "benign and,many would say, beneficial speech," and also by asserting that the possibility thatdetailing might persuade "provides no lawful basis for quieting it." '294

iii. Direct Advancement of Interest?

Although the Sorrell majority narrowly defined the government interest, and thuseasily rebutted why these interests were not advanced, it still considered whether thegovernment interest of reducing brand name drug costs was directly advanced-allwhile being influenced by the schemas. However, as the dissent pointed out, thechallenged state law did directly advance a government interest by ensuringdiscussion is based on actual facts about drugs, rather than using past doctorprescriptions to sway doctors.2 9

The Sorrell majority repeatedly relied on schemas to support its assertion that theregulation did not directly advance any government interest. For example, themajority noted that doctors find detailing "instructive"--consistent with thesophisticated doctor schema, as well as the more information schema.2 96 The courtalso noted that the common First Amendment refrain that truthful information shouldnot be suppressed based on the fear of bad decisions (i.e., the more informationschema) applies with "full force" when the audience consists of "sophisticated andexperienced consumers.2 97 In doing so, the majority quoted Edenfield, a case thatfound a law banning certified public accountants (CPAs) from personal solicitationto be unconstitutional since prospective clients were likely sophisticated and

289. Sorrellv. IMS Health Inc., 564 U.S. 552, 572-74 (2011).290. Id. at 575.291. Id. at575-77.292. Id. at 576-78.293. Id. at 597-98 (Breyer, J., dissenting) (finding a broader substantial interest to regulate

public health consistent with police powers).294. Id. at 576 (citing Brandenburgv. Ohio, 395 U.S. 444, 447 (1969)) (noting that fear of

persuasion is not a lawful basis for quieting speech in most situations, including this case).295. Id. at 597 (Breyer, J., dissenting).296. Id. at 578.297. Id. at 577 (emphasis added) (quoting Edenfield v. Fane, 507 U.S. 761, 775 (1993)).

[Vol. 94:773


experienced business executives not prone to manipulation.29' However, that case

found no dangers for professionals subject to in-person solicitation by a CPA who isnot trained in the "art of persuasion. ",299 Although doctors are professionals, theyare being approached by pharmaceutical representatives whose livelihoodfundamentally depends on being persuasive.3"' In fact, companies are known to hirethose they consider most likely to be persuasive.3"1

The schemas may help explain some of the discussion in the Caronia majorityopinion concerning the direct advancement of interest. Although the majorityrecognized that the government has a substantial interest in preserving the integrityof the drug approval process and reducing patient harm from unsafe and ineffectivedrugs, its reliance on the information schema seemed to unduly contribute to itsconclusion that a bar on off-label marketing of "truthful" information fails to directlyadvance this interest.30 2 In particular, the majority's assertion that FDA regulationslegalize off-label use, but "prohibit[] the free flow of information that would informthat outcome,"3 3 reflects the more information schema in conjunction with thesophisticated doctor schema that falsely presumes doctors are not misled bymarketing. So, the majority's assertion that drug companies were providing"potentially relevant treatment information" that will provide "informed andintelligent treatment decisions" does not comport with reality.30 4 There are alsosystematic problems with relying on the schemas. As recognized by the dissent,permitting a company to promote a drug for any use after the drug has been onlyapproved for one would fundamentally disrupt the entire process for drug approvaland portends that the entire process could be found unconstitutional.3 5

iv. Narrowly Tailored

In addition, the Caronia majority's suggestion that prosecution of off-labelmarketing is not narrowly tailored is also questionable once cognitive biases aretaken into consideration.3 6 For example, the majority suggested that a less speech-

298. Id. In addition, in Edenfield, the Court noted that CPA training emphasizesindependence and objectivity, not advocacy, which are likely inapplicable to drugrepresentatives. See Edenfield, 507 U.S. at 775.

299. Id. (emphasis added).300. Not only are pharmaceutical representatives hired to be persuasive, but about a quarter

of their overall compensation is tied to prescriptions written by physicians. See Christopher v.SmithKlein Beecham Corp., No. CV-08-1498-PHX-FJM, 2009 WL 4051075, at *3 (D. Ariz.Nov. 20, 2009).

301. E.g., Stephanie Saul, Gimme an Rx! Cheerleaders Pep Up Drug Sales, N.Y. TIMES(Nov. 28, 2005), https://www.nytimes.com/2005/11/28/business/gimme-an-rx-cheerleaders-pep-up-drug-sales.html [https://perma.cc/49YU-88VG].

302. See United States v. Caronia, 703 F.3d 149, 166 (2d Cir. 2012).303. Id. at 167.304. Id. at 166.305. See id. at 179.306. In contrast, the Caronia dissent seemed to implicitly understand that more

information is not always better in considering misbranding to be the least restrictive way toadvance the government interest in protecting the new drug approval process, which theSupreme Court has previously recognized as an important government interest. Id. at 178.

2019]

INDIANA LAW JOURNAL

restrictive alternative could be to educate doctors regarding potentially misleadinginformation.31

7 However, this suggestion is questionable based on what studies haveshown: doctors are already skeptical of the accuracy of information frompharmaceutical companies, and yet are still vulnerable to the marketing messages farmore than they realize.31

8 In addition, as discussed earlier, it is difficult to changepreexisting schemas due to confirmation bias that impacts all individuals.3 9 That is,if a doctor believes that she is capable of distinguishing misleading information fromtrue information-even if she is in fact incapable of doing so-she is likely to ignoresubsequent information inconsistent with her beliefs. So, simply telling doctors thatthey are vulnerable to being misled is not likely to be adequate alone.

The information schema also seemed highly relevant to the Amarin decision thatfollowed Caronia. The court noted that it wanted to "err on the side of caution,"which meant providing doctors with more, not less, information, consistent with themore information schema.310 Since the plaintiff-doctors alleged the information washelpful, the court did not question this belief, even though, as previously discussed,studies show doctors are in fact far less sophisticated than they realize.311

The more information schema is reflected in the Amarin court's evaluation andapproval of a proposed statement regarding how certain omega-3 fatty acids, presentin Amarin's drug marketed as Vascepa, may reduce heart disease according to"[s]upportive but not conclusive research.31 2 Although the FDA considered this tobe potentially misleading, the court focused on the fact that the FDA did notchallenge the claim as completely misleading.313 In other words, the court's lack ofconcern regarding potentially misleading information is consistent with the moreinformation schema that does not question harm from potentially misleadinginformation.314 The court noted that the accompanying disclosures would easilyaddress any such problem. 315 However, as explained in Section II.C, relying ondisclosures to actually inform is not supported by empirical studies.

307. Id. at 168.308. See supra Section II.B.2.309. See supra Section II.A.2.310. AmarinPharma, Inc. v. FDA, 119 F. Supp. 3d 196, 233 (S.D.N.Y. 2015).311. See supra Section II.B.

312. Amarin, 119 F. Supp. 3d at 214. Amarin proposed the following Statement (#1):"Supportive but not conclusive research shows that consumption of EPA and DHA omega-3fatty acids may reduce the risk of coronary heart disease." Id. at 234. Similarly, the informationschema likely led the court to permit Amarin to add additional language to the FDA's proposedstatement regarding the fact that recent trials failed to demonstrate the cardiovascular benefitof adding a second lipid-altering drug when it characterized the industry desired language as"factually accurate," and relegated to a footnote its dismissal of the FDA's assertion that thelanguage was misleading. Id. at 233 n.65.

313. Id. at235.314. See id. at 234.315. Id. at 235 (noting FDA concern that a doctor might improperly conclude that "there

is currently sufficient evidence to support a conclusion that drug-induced decreases intriglyceride levels lead to a reduction in the risk of cardiovascular events in patients on statintherapy."). The other agreed upon disclosures are that the "FDA has not approved Vascepa toreduce the risk of coronary heart disease" (Amarin Disclosure #1); " [ t]he effect of Vascepa onthe risk of cardiovascular mortality and morbidity has not been determined" (Amarin

[Vol. 94:773


3. Schemas Baked into Precedent Result in FDA Capitulation

The two schemas are currently entrenched in commercial speech precedent in partas a result of a lack of government challenges to prior rulings. Although the lack ofgovernment challenges to prior rulings can be considered reasonable to avoidpotential court decisions that could threaten the entire FDA new drug-approvalprocess, government inaction could ironically lead to the same conclusion. To bestunderstand the current situation, it is important to begin with the first case to use theFirst Amendment to challenge FDA regulation of off-label uses (i.e., WashingtonLegal Foundation) since the FDA's response to that has impacted subsequent cases.

Washington Legal Foundation was a key case in finding FDA regulations barringcompanies from distributing peer-reviewed articles concerning off-label uses toviolate commercial speech. After losing the summary judgment motion inWashington Legal Foundation, the FDA effectively conceded companies coulddistribute peer-reviewed articles that support off-label uses.316 However, empiricalstudies conducted since the case was decided in 1998 now show serious problemswith peer-reviewed articles, even if they have the veneer of objectivity.3 17 But,because the FDA settled, subsequent courts have used this decision to suggest thatthe FDA itself acknowledges that such articles are helpful.318

Disclosure #3); "[a] cardiovascular outcomes study of Vascepa designed to evaluate theefficacy of Vascepa in reducing cardiovascular mortality and morbidity in a high risk patientpopulation on statin therapy is currently underway" (Amarin Disclosure #4); and "Vascepamay not be eligible for reimbursement under government healthcare programs ... to reducethe risk of coronary heart disease .... (Amarin Disclosure #5); as well as contested AmarinDisclosure #2, which stated that the "FDA has not approved Vascepa for the treatment ofstatin-treated patients with mixed dyslipidemia and high... triglyceride levels."

316. See U.S. FOOD & DRUG ADMIN., GOOD REPRINT PRACTICES FOR THE DISTRIBUTION OF

MEDICAL JOURNAL ARTICLES AND MEDICAL OR SCIENTIFIC REFERENCE PUBLICATIONS ON

UNAPPROVED NEW USES OF APPROVED DRUGS AND APPROVED OR CLEARED MEDICAL DEVICES

(2009), https://www.fda.gov/regulatoryinformation/guidances/ucml25126.htm [https://perma.cc/K5QV-4SG5]. The details that led to this are a bit complex. Basically, before the districtcourt ruling in favor of the Washington Legal Foundation (WLF), Congress passed the Foodand Drug Administration Modernization Act of 1997 (FDAMA), which permitted informationabout unapproved uses so long as the manufacturer had applied or intended to apply for a newuse. Pub. L. No. 105-115, 111 Stat. 2296 (codified as amended at 21 U.S.C. § 360). Althoughthe FDA initially thought the WLF ruling enjoining its previously challenged guidancedocuments did not apply to the new statute, after the district court enjoined enforcement ofFDAMA and its implementing regulations, the government initially appealed, but reached anagreement with WLF that it had no independent prosecutorial authority, only the ability toprovide a "safe harbor" from prosecution. Wash. Legal Found. v. Henney, 202 F.3d 331, 335-36 (D.C. Cir. 2000); see also Krause, supra note 83. In addition, although FDAMA provisionswere written to sunset in 2006, the 2009 FDA guidance document was more beneficial tocompanies because they could provide documents concerning off-label uses without any needto have an intent to submit an application to the FDA for approval of that use. Krause, supranote 83, at 410.

317. See supra notes 59-60 and accompanying text.318. See United States v. Caronia, 703 F.3d 149, 166-67 (2d Cir. 2012).

2019]

INDIANA LAW JOURNAL

More recently, the Amarin court recognized that there was a "vigorous dissent"in Caronia but also noted that the FDA failed to challenge this decision, seeming tosuggest that this means that the Caronia majority was necessarily correct.319

Although many commentators recognize that the FDA's failure to challengejudgments and enter into settlements is inherently tied to concern about furthererosion of its regulatory powers, its inaction has opened the door to courtsquestioning its basic regulatory authority. For example, the Amarin court assertedthat the FDA process for approving even new drugs "predates modem [sic] FirstAmendment law respecting commercial speech.'32" The FDA did not challenge this,likely because an appeal would have been futile since the case was filed in theSouthern District of New York and governed by Caronia.321 Nonetheless, the Amarinsuggestion is a troubling one that subsequent courts can now rely on. In addition, thistroubling suggestion may result in further FDA capitulation to settlements in othercases.

B. The Mismatch Between First Amendment Policy and Pharmaceutical Marketing

There is currently a stark mismatch between First Amendment policy andmarketing of prescription drugs. Although commercial speech cases consistentlyassert that well-informed decisions are a goal, this is often an elusive one, withdangerous public health consequences looming in the context of promotingunapproved uses of prescription drugs. As previously discussed, it is difficult fordoctors to make informed prescription decisions when they rely on information fromself-interested companies. This Section explains that although case law recognizesmisleading commercial speech as raising policy concerns that justify moreregulation, cases thus far fail to see that many of these concerns exist in the contextof potentially misleading speech from pharmaceutical companies that speak in a one-sided marketplace. This Section explains why recognition of these similarities isimportant to achieve the policy goal of informed decisions.

1. Policy Similarities Between Inherently and Potentially Misleading Speech

Although it is undisputed that inherently misleading commercial speech may bebarred, the policy concerns behind such speech are actually similar to those ofpotentially misleading speech. For example, the Amarin court stated that "[w]hetherspeech is 'inherently misleading' depends on ... the 'possibilities for deception,'..whether 'experience has proved that in fact that such advertising is subject to abuse,'

. and 'the ability of the intended audience to evaluate the claims made.' 322

319. Amarin, 119 F. Supp. 3d at 227.320. Id. at 226.321. The FDA promptly settled after the court's holding in favor of granting a preliminary

injunction against the FDA, essentially permitting the company to assert their desired claims.See Eric Palmer, With FDA Settlement, Tiny Amarin Creates Opening for Pharma in Off-LabelMarketing, FWRCEPHARMA (Mar. 9, 2016, 10:10 AM), https://www.fiercepharma.com/pharma/fda-settlement-tiny-amarin-creates-opening-for-pharma-off-label-marketing [https:/perma.cc/KS9X-72AY].

322. Amarin, 119 F. Supp. 3d at 235 (quoting Wash. Legal Found. v. Friedman, 13 F. Supp.

[Vol. 94:773


Although not recognized by courts thus far due to the two schemas, doctors, who arethe intended audience of pharmaceutical marketing, have no ability to verifyaccuracy of pharmaceutical marketing since there is no other source of information.In addition, although not yet recognized by courts, studies show that doctors arevulnerable to advertising, yet are largely unaware of their vulnerability.

Detailing has characteristics similar to those previously noted as problematic incommercial speech cases regarding misleading information that can be banned. Inparticular, the Supreme Court has previously given states more liberty to restrictspeech in situations where there is a danger of overreaching. This can happen whensomeone who is "trained in the art of persuasion" is in a situation that would breed"undue influence," such as an attorney who is soliciting at an accident site or hospitalroom.3 23 As noted earlier, sales representatives are hired to be persuasive, and theirlivelihood depends on it.3 24 Studies show that these representatives have undueinfluence on doctors based not on a superior product, but on their effective use ofmarketing techniques known to take advantage of psychological biases. Indeed, thefact that many doctors claim the information from representatives is useful withoutimpacting them-contrary to actual studies from evidence-shows that they aresubject to influence. Whereas there is general consumer skepticism concerningattorneys that might make even an accident victim leery of such a solicitation,doctors' assertions of the value of pharmaceutical detailing underscore thatrepresentatives are so persuasive that doctors are largely unaware that they are beingmanipulated.

Also, the Supreme Court has noted that deception is more likely when there is alimited ability to self-police and a lack of standardization in the product.3 25 Althoughthese comments were made about attorney advertisements (where the product isattorney advice, which is usually individualized), the absence of self-policing and astandard product apply equally to the field of pharmaceutical marketing. 326 It is clearthat the industry is not self-policing its own conduct considering a long history ofcompanies being fined for improperly advertising off-label use of drugs.3 27 Inaddition, there is no standardization for off-label drug marketing; the FDA onlyregulates the uses that it has evaluated, not off-label uses.3 28 Of course, the case that

2d 51, 67 (D.D.C. 1998) (internal citations omitted)).323. Ohralikv. Ohio State Bar Ass'n, 436 U.S. 447, 461, 464-65 (1978).324. See supra notes 300-301 and accompanying text.325. See In re R.M.J., 455 U.S. 191, 202 (1982) (relying extensively on Bates v. State Bar

of Ariz., 433 U.S. 350 (1977), to establish that there is no absolute bar on potentiallymisleading information and that the attorney advertising in case at issue was not actuallymisleading to include information about where the attorney was licensed to practice or to mailcards announcing the opening of an office).

326. In addition, they are arguably even more concerning in the context of off-labelmarketing since there are public health harms from misleading drug advertisements butgenerally not regarding attorney advertisements.

327. See, e.g., supra notes 84-86 and accompanying text; see also Amarin, 119 F. Supp.3d at 204 ("There are many examples in which prescriptions of an approved drug for off-labeluse has caused harm.").

328. See 21 C.F.R. § 201.57(a)(3) (2015) (requiring labeling only for approved uses ofdrugs). The FDA regulates labeling of approved uses of drugs, but does not directly regulateunapproved uses. Id.; see also supra Section L.A (explaining that the FDA approves drugs for

2019]

INDIANA LAW JOURNAL

mentioned these issues also noted the public's comparative lack of knowledge as akey factor, which would seem initially inapplicable to doctors.32 9 However, asexplained in Part II, although doctors are highly educated, they are far lesssophisticated than the sophisticated doctor schema assumes, and studies show thatthey actually have been misled. So, rather than deception being simply a theoreticalissue that courts readily dismiss as improbable, it is in fact often the reality.

2. Potentially Misleading Pharmaceutical Marketing Is Unlikely to BeCorrected, with Significant Public Harm, but Limited Benefit

An important issue with marketing of prescription drugs is that the traditionalpresumptions concerning how potentially misleading speech will be corrected arefalse. In particular, as previously discussed, cases traditionally assume thatpotentially misleading speech will be corrected either by another source of speech,or, alternatively, by a disclaimer. However, as will be explained, neither can combatalleged facts presented by a self-interested company in the unique prescription drugmarketplace. In addition, although commercial speech case law generally does notfocus on how persuasive the speech is, that presumption may be less justified in theunique circumstance of off-label drug promotion when there is no contrary speechand the potential public health harm is severe.

Additional information, whether from a source other than the self-interestedcompany or in the form of a disclaimer, is unlikely to make potentially misleadinginformation from such companies no longer misleading. As noted earlier, disclaimersare ineffective at minimizing confusion, even for well-educated audiences such asscientists. In addition, when a new drug is first introduced, and for a number of yearsafter that, there is only one company that has information to speak about this drug.33

Accordingly, unlike the typical market that commercial speech law presumes, wherethere can be a marketplace of different ideas, the abnormal pharmaceutical marketunduly represents the views of the industry. Furthermore, as noted in Part II, doctorsare vulnerable to being misled by pharmaceutical marketing-and may continue toimproperly rely on marketing even in the face of contrary evidence. This is consistentwith the fact that studies show repeated information is more likely to be consideredtrue.331 Moreover, independent research typically results in peer-reviewed articlesthat are inherently less accessible than detailing and also not as likely to be repeatedas commercial marketing messages. Subsequent independent information will alsobe considered less due to the cognitive bias of anchoring, whereby initial informationtends to be the basis upon which subsequent information is perceived. So, any initialmisinformation is not truly corrected by another source of speech.

In addition, the potential to mislead a doctor into prescribing a drug for anunapproved use has serious public health consequences but little of the traditional

specific uses); supra Section I.C (noting that there is no direct regulation of off-label uses, but,rather, liability for introducing misbranded drugs based on any indication of drug use that hasnot been FDA approved).

329. InreR.L.J., 455 U.S. at 202.330. This is likely since new drugs are typically patented, and patent rights give their owner

the right to exclude all others from making or selling the same thing during the term of a patent.331. See supra text accompanying notes 145-147.

[Vol. 94:773


First Amendment benefits with respect to promoting democratic principles.33 2

Commercial marketing of drugs is unlikely to promote debate on public policy. Thesole goal of drug marketing is to increase sales. Since there is already extensiveskepticism concerning the value of commercial marketing for uses of drugs that areentirely FDA approved, marketing of uses not approved by the FDA should becautiously considered since these are associated with more adverse effects. As notedearlier, the potential harms may be serious. Although government interests play arole in evaluation of the Central Hudson factors for potentially misleading speech,recognizing that stated policy concerns about misleading speech have somesimilarities with the area of prescription drug promotion is also important inconsidering how to address issues.

IV. AN INFORMED APPROACH TO DRUG ADVERTISING AND DEVELOPMENT

This Part proposes possible solutions to directly address the implications of thepreviously unveiled and debunked schemas. Section A focuses on legal steps tocorrect the current undue influence of schemas in commercial speech jurisprudence.Section B goes beyond the legal realm to suggest structural changes to minimize theimpact of schemas.

A. Aligning First Amendment Law and Policy with Reality

This Section proposes several changes to commercial speech law to betterrecognize that the current lopsided marketing of drugs to doctors results in their beingmisled, even though they are dangerously unaware of this. There are four specificchanges to the law that are recommended. First, courts should recognize that"potentially" misleading information is different than "truthful and non-misleading"information and treat it differently as a matter of law in evaluating commercialspeech; specific factors are provided to guide courts on how to provide an appropriate

332. See Rosenberger v. Rector & Visitors of Univ. of Va., 515 U.S. 819, 831 (1995);Hustler Magazine v. Falwell, 485 U.S. 46, 55 (1988); see also TAMARA R. PIETY,

BRANDISHING THE FIRST AMENDMENT: COMMERCIAL EXPRESSION IN AMERICA 165-85 (2012)

(raising issues concerning protection of commercial speech in the context of promotingdemocracy); Robert Post, The Constitutional Status of Commercial Speech, 48 UCLA L. REV.1, 4, 49 (2000) (noting that whereas the First Amendment bars the state from suppressingpublic discourse due to democratic governance, this policy issue does not apply to commercialspeech that is intended instead to focus on promoting information); supra notes 100-109 andaccompanying text (explaining different treatment of commercial speech versusnoncommercial speech). Of course, there are some that believe that commercial speech cannonetheless be tied to democratic self-governance. E.g., Martin H. Redish, CommercialSpeech, FirstAmendment Intuitionism and the Twilight Zone ofViewpoint Discrimination, 41Loy. L.A. L. REV. 67, 81 (2007) ([S]peech concerning commercial products and services canfacilitate private self-government in much the same way that political speech fosters collectiveself-government."); Daniel E. Troy, Advertising: Not "Low Value" Speech, 16 YALE J. ONREG. 85 (1999) (challenging the notion that commercial information or advertising is lessvaluable than other forms of speech).

2019]

INDIANA LAW JOURNAL

balance between promoting First Amendment protections while still recognizing thepossible harms of such information. Second, the burden of proof should be shiftedregarding disclaimers; rather than have the FDA prove that information would bemisleading, the burden should be on the company to establish that a disclaimer willnot be misleading. Third, FDA guidance should be reevaluated and informed bycognitive biases. Fourth, courts should defer more to FDA decisions regardingwhether information is potentially misleading.

1. An Informed Treatment of Potentially Misleading Information

Whereas current First Amendment law treats completely truthful and potentiallymisleading information as equivalent as a matter of law, these two types ofinformation should be legally distinct-and treated as such when evaluating theCentral Hudson factors.333 It is a legal fiction that completely truthful and potentiallymisleading information are the same. Accordingly, if information is potentiallymisleading, the government should be given more flexibility in regulating that speechsince it has a higher likelihood to mislead. In addition, courts should consider thatthe FDA has a substantial interest in regulating potentially misleading informationthat has important public health consequences. This is particularly true in the case ofoff-label uses that are often not scientifically supported and can result in negativepublic health outcomes.

Embracing a different treatment for potentially misleading information wouldhelp to avoid oddities like Caronia where not only was the information not truthful,but the court relied on statements in prior case law that seem questionable with regardto information that is potentially misleading. For example, the Caronia majority citedprior Supreme Court precedent that a ban on "truthful and non-misleading" speechrests "solely on the offensive assumption that the public will respond 'irrationally'to the truth.' 334 While it may seem "offensive" that individuals respond irrationallyto the truth, cognitive bias studies consistently indicate that individuals often do notrespond rationally.335 In addition, pharmaceutical marketing favorsmisrepresentation that is necessarily potentially misleading. Unlike othercommercial speech challenges involving facts that can be easily verified, such asadvertised prices, off-label marketing is a situation where the alleged facts cannot beverified since the only source of information is the self-interested company.336

333. Given the cognitive biases at issue and serious public health risks implicated, therecould be an argument for presuming that all marketing of off-label uses should be presumedactually misleading unless proven otherwise. After all, pharmaceutical marketing ofteninvolves alleged facts for which a court cannot readily assess validity. However, recognizingthat this would seem to give the government a complete free pass to bar speech that runs totallycontrary to First Amendment jurisprudence. This Article instead makes the more modestsuggestion that potentially misleading information be viewed more skeptically.

334. United States v. Caronia, 703 F.3d 149, 166-67 (2d Cir. 2012) (quoting Liquormart,Inc. v. Rhode Island, 517 U.S. 484, 503 (1996) (striking down ban on price advertisement ofalcohol)).

335. See supra Section II.A.2 (explaining a number of cognitive biases that result inindividuals maintain incorrect beliefs).

336. Moreover, even with an advertisement about prices for attorney services, not all

[Vol. 94:773


Treating potentially misleading information differently than actually truthful andnon-misleading information would likely also have been beneficial in Amarin. Thecourt stated that the FDA "argues only that the claim is 'potentially misleading,"' asif there are no possible harms with potentially misleading doctors on an issue thatimpacts public health.337 Of course, the statement is consistent with current case law.However, some have questioned whether a key claim the court considered truthfuland unlikely to mislead even had any factual basis.338 In particular, the claim thatthere is "supportive but not conclusive research" that the drug at issue "may reducethe risk of coronary heart disease" has been suggested as equivalent to saying that it"may or may not reduce the risk."339 In other words, Amarin's statement is "true" inthe sense that it has relatively little factual content such that its content is notaffirmatively false. Nonetheless, it could still be potentially misleading, which thecourt did not actually consider since it did not need to under current law that treatspotentially misleading information as equivalent to completely nonmisleading andtruthful information. However, given the lack of actual sophistication of doctors,there is a potential to mislead.

Of course, the question is how a court should assess whether commercial speechhas the potential to mislead. There are two key factors that courts can and shouldconsider whether information is potentially misleading: (1) whether the commercialinformation can be verified by at least one other source, and especially whether thatsource is objective and reliable; and (2) the societal cost of misinformation, includingthe scope of public health harm. The existence of at least one should be consideredby a court in coming to the legal determination that information is potentiallymisleading. Each of these factors will be briefly explained.

An initial issue is whether the commercial speech can be verified by at least oneother source, and whether such source(s) are objective and reliable. Importantly, it isessential to look beyond merely whether the source is a peer-reviewed journal articlebecause of previously noted problems with the peer-reviewed articles, including thatthe industry may have undue influence with these articles. In other words, contraryto prior cases, courts should not assume that a peer-reviewed journal article isobjective and unbiased. In addition, sometimes peer-reviewed journals will not evenbe necessary to find that the commercial information is not accurate and thus has thepotential to mislead. For example, in Caronia, there was contrary evidence to thecommercial speech concerning off-label use of the strongest type of weight-anFDA black box warning specifically cautioning against the proposed use.340

justices agreed that such advertisements are not misleading. See Bates v. State Bar of Ariz.,433 U.S. 350, 386 (1977) (Burger, C.J., concurring in part, and dissenting in part) (notingskepticism that legal services can be generally considered " standardized" such that advertisingis not misleading); id. at 391 (Powell & Stewart, JJ., concurring in part, and dissenting in part)(noting that advertising of legal services has "long been thought ... inevitably ... misleadingbecause such services are individualized").

337. AmarinPharma, Inc. v. FDA, 119 F. Supp. 3d 196, 235 (S.D.N.Y. 2015).338. See SpencerPhillips Hey & Aaron S. Kesselheim, An Uninformative Truth: The Logic

ofAmarin's Off-Label Promotion, PLoS MED., Mar. 2016, at 2.339. Id.340. Caronia, 703 F.3d 149 at 155.

2019]

INDIANA LAW JOURNAL

Another important issue is the public health and societal cost of themisinformation. For example, in Caronia, encouraging off-label use that is subjectto a black box warning could have tragic consequences that should have strongweight in concluding information is potentially misleading. However, this factorshould not be limited to considering only tragic consequences. If the commercialspeech would encourage avoiding other treatments, especially ones known to beeffective and less expensive, that would also impose a societal cost in terms ofunnecessary expense. So, for example, with a drug that would not likely result inserious public health harm, such as the purified fish oil sold by Amarin, the desiredcommercial speech concerning an unapproved use could still be consideredpotentially misleading if it would impose unnecessary costs for the drug withoutsubstantial benefit.

2. A Modified Burden on Disclaimers

An important change to the law that is well supported by empirical data is tofundamentally change the burden of proof regarding disclaimers before consideringthat a disclaimer is a less restrictive alternative to barring potentially misleadingspeech. As discussed earlier, courts currently presume that disclaimers will beeffective in minimizing confusion and place the burden on the party seeking torestrict speech, such as the FDA, to show that the disclaimer will confuse. Moreover,some courts require that this showing be grounded in empirical evidence.341

However, empirical evidence shows that disclaimers actually are not understood byconsumers.34 2 Case law should comport with actual evidence of how consumersprocess, or, more appropriately, fail to process disclaimers.34 3 In particular, drugcompanies should have the burden of establishing why a disclaimer would beeffective in minimizing the possibility of being misled before a court can considerthat as a viable less speech-restrictive alternative. If a company cannot establish thata disclaimer will actually make its proposed speech less confusing or misleading,then a court can more easily find government regulation, such as barring off-labelmarketing, to be narrowly tailored since a disclaimer would not be a true alternative.Admittedly, permitting a company to speak with a disclaimer is always less speechrestricting, but the current presumption that a disclaimer is effective at minimizingconfusion is flawed.

The proposed burden shift would help address oddities in prior cases. Forexample, in Caronia, the company would need to explain how a disclaimer wouldadequately protect doctors from being misled. In other words, Caronia would beforced to argue what type of disclaimer could guard against its representativeasserting that off-label use was not harmful for an elderly patient when that exact use

341. See, e.g., Whitaker v. Thompson, 248 F. Supp. 2d 1, 13 (D.D.C. 2002).342. See supra Section II.C.2.343. In the context of trademark infringement, courts have sometimes recognized that

consumers do not necessarily read disclaimers based on similar studies based on consumerpsychology. See Jacob Jacoby & Robert Lloyd Raskopf, Disclaimers in TrademarkInftingement Litigation: More Trouble than TheyAre Worth?, 76 TRADEMARKREP. 35 (1986);Gita Venkataramani Johar & Carolyn J. Simmons, The Use of Concurrent Disclosures toCorrect Invalid Inferences, 26 J. CONSUMER RES. 307 (2000).

[Vol. 94:773


was subject to a black box warning. Not only does this seem like a tall order, butgiven prior studies showing that doctors are not always aware of black box warningson other products, it is questionable that the black box warning or any otherdisclaimer could counteract the more easily accessible information provided indetailing.

3. More Judicial Deference to FDA Evaluation

Courts should also be more deferential to FDA determinations concerningwhether scientific information is either misleading or potentially misleading, or atleast no longer require the FDA to provide empirical evidence that a disclaimer willconfuse consumers. As noted earlier, there is extensive literature showing thatdisclaimers are ineffective. This is obviously relevant to any litigation regardingmarketing of off-label uses, as well as marketing of products to consumers such asdietary supplements.344 This is not to suggest that courts should accept FDAassertions at face value without any inquiry. However, the current standards seempremised on the more information schema. For example, courts have suggested thatinformation can only be banned as actually misleading if there is no empirical supportat all, or alternatively, only one to two studies that support a claim in addition toempirical evidence that an additional disclaimer would still result in consumerconfusion.345 This might be relevant to whether information is actually misleading,but information could be potentially misleading if only a minority of studies, even ifmore than one to two in total, support a claim.

Although courts routinely evaluate scientific information in general, these casesshow that courts are vulnerable to the more information and sophisticated doctorschemas. In contrast, the FDA has not been vulnerable to these positions. To thecontrary, the FDA's original position of limiting companies from distributing peer-reviewed articles to doctors for fear of the potential to mislead shows that the FDAdoes not embrace either position. And, as discussed earlier, the FDA's originalposition is now empirically supported. On some level, courts recognize that they areless equipped to evaluate key information. For example, although the judge inAmarin evaluated a number of statements for alleged truthfulness, at the actualhearing he admitted: "You're talking to somebody who has difficulty using a toaster

... I'm the last person who should opine on this." 346 In addition, a judge in a single

344. Of course, the context in which the FDA may be opining on whether information ispotentially or actually confusing will be different. For example, since the FDA does not reviewoff-label uses, it is only in the context of a First Amendment challenge, such as Amarin, thatthis would be likely. In contrast, the FDA regularly opines on whether claims concerningdietary supplements and food products are potentially confusing when manufacturers of suchproducts submit proposed health claims for FDA review to ensure that there is "significantscientific agreement" in support of the claim. See 21 U.S.C. § 343(r)(3)(B)(i) (2012) (healthclaims on food); id. § 343(r)(5)(D) (health claims on dietary supplements); see also 21 C.F.R.§ 101.14(c) (2017) (standard of FDA approval premised on totality of publicly availablescientific evidence and significant agreement among experts that there is evidence in supportof the claim); Id. § 101. 14(a)(1) (defining health claim that requires FDA approval).

345. E.g., Whitaker, 248 F. Supp. 2d at 10.346. Amy Kapczynski, Free Speech and Pharmaceutical Regulation Fishy Business, 176

2019]

INDIANA LAW JOURNAL

case will not have the same depth of experience as the FDA and could be persuadedby information that may not be scientifically valid.

Recognizing and respecting FDA expertise is an important and timely issue givenproposals to remove traditional judicial deference to all agency determinationsgenerally, which would also impact the FDA. In particular, in 2017, the House passeda bill to overturn the common-law principle of "Chevron deference ,

' 31

7 which arosefrom the 1984 Supreme Court case, Chevron v. Natural Resources DefenseCouncil.348 Under the Chevron deference, if a law passed by Congress is silent orambiguous with respect to an issue, courts should defer to agency interpretationunless it is unreasonable (i.e., "arbitrary, capricious, or manifestly contrary to thestatute").34 9 Accordingly, courts are not to substitute their view for that of the agency,even if another interpretation is reasonable. Although the legislation applies toagencies' interpretations of all statutes, and has already been criticized, the schemasrevealed in this article provide additional report for opposing such legislation, whichis currently under consideration by the Senate .3" This legislation would make thesituation even worse than it currently is.

4. Reevaluating FDA Guidance

In addition, the FDA should carefully consider existing cognitive biases if itconsiders further revising its existing guidelines regarding pharmaceutical marketingof off-label uses. The FDA has already relaxed the guidelines in light of pressureafter judicial decisions such as Caronia and Amarin.351 However, although the FDA

J. AM. MED. ASS'N INTERNAL. MED. 295, 295 (2016) (quoting Transcript of PreliminaryInjunction Hearing at 79, Amarin Pharma, Inc. v. FDA, 119 F. Supp. 3d 196 (S.D.N.Y. 2015)(No. 15 Civ.V 3588 (PAE))).

347. H.R. 5, 115th Cong. (2017-2018). In addition, it passed through the committee phaseof the Senate. S. 951, 115th Cong. (2017).

348. Chevron, U.S.A. Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837 (1984).349. Id. at 844; see also Thomas A. Lorenzen & Sharmistha Das, The Decline of

Deference: Is the Supreme Court Pruning Back the Chevron Doctrine?, TRENDS: ABA SEC.OF ENV'T, ENERGY, & RESOURCES NEWSL., Sept./Oct. 2015.

350. For critiques of proposed legislation, see, for example, Jonathan J. Darrow, Erin C.Fuse Brown & Aaron S. Kesselheim, The Regulatory Accountability Act of 2017Implications for FDA Regulation and Public Health, 378 NEW ENG. J. MED. 412 (2018);William Funk, Requiring Formal Rulemaking Is a Thinly Veiled Attempt to Halt Regulation,REG. REV. (May 18, 2017), https://www.theregreview.org/2017/05/18/funk-formal-rulemaking-halt-regulation [https://perma.cc/TJH7-ANWJ].

351. In June 2018, the FDA issued two guidance documents that expand the scope ofpermissible corporate communications beyond its proposed 2017 guidelines. See supra note81. The guidance documents were long anticipated. See, e.g., GIBSON DUNN, 2017 YEAR-END

FDA AND HEALTH CARE COMPLIANCE AND ENFORCEMENT UPDATE - DRUGS AND DEVICES

(2018) (noting that the FDA had not provided updated guidance as of January 2018 after twoyears of anticipation); ELIZABETH RICHARDSON, HEALTH AFFAIRS, HEALTH POLICY BRIEF:

OFF-LABEL DRUG PROMOTION. DRUG COMPANIES ARE LARGELY PROHIBITED FROM

PROMOTING A DRUG FOR USES THAT HAVE NOT BEEN APPROVED BY THE FOOD AND DRUG

ADMINISTRATION (2016) (noting that the FDA had announced that it would be providing newguidance in 2016).

[Vol. 94:773


guidelines may seem consistent with these decisions, even the prior 2014 guidelinesconcerning the safe harbor from prosecution for misbranding permitted action thatcognitive bias studies suggest exposes doctors to undue influence. For example, since2014, after the FDA capitulated to a prior challenge, companies can share peer-reviewed articles supporting off-label use with doctors contrary to long-standingFDA concerns that recent studies show is now justified;35 2 in particular, there arerecognized problems with the publication process that tend to favor publication ofpositive results-even before considering that drug companies may be selectivelypresenting only supportive articles.353 In addition, even though doctors know at thetime that companies show them articles that are likely self-selected, the articles maystill be unduly persuasive. Over time, doctors may forget what information theylearned from the biased source and simply remember the information that getsrepeated, as it may be through multiple articles, and erroneously assume that it istrue."' In other words, contrary to the 2014 FDA guidance documents permittingcompanies to share peer-reviewed articles on off-label uses, the FDA's originalinstinct to bar companies from sharing such information for fear of undue influenceof unsupported information was correct. It seems unlikely that the FDA will retreatfrom the increasingly expanded safe harbor for off-label marketing practices.Nonetheless, recognizing that existing case law may be inconsistent with empiricalevidence should give the FDA pause before further capitulating to industry demands.

B. Suggestions for Structural Reforms

This Section sketches suggestions for structural reforms to address the issuesraised. This Section first suggests increasing awareness of the extent andeffectiveness of marketing influence to minimize the sophisticated doctor schema.This Section also proposes specific actions to limit marketing influences from self-interested companies by both limiting the extent of marketing, as well as promotingindependent and accessible information. Finally, this Section argues that the datarevealed here support more extensive changes to the current system of drugdevelopment reliant on profit-based companies.

1. Increase Awareness

To combat the sophisticated doctor schema, an important initial step is to increaseawareness of marketing influences on doctors. Lack of awareness is likely the reasonmost doctors believe they are not adversely influenced by detailing and improperlyassume they are able to glean relevant information.355 This Section focuses primarilyon educating doctors who are being targeted with advertisements and for whom thereare existing channels to provide such information. Judges need to be informed as

352. For example, the Caronia majority quoted from the FDA's own guidelines that theFDA recognized the value of truthful and nonmiisleading scientific information on unapproveduses, such as through scientific journals. United States v. Caronia, 703 F.3d 149, 166-67 (2dCir. 2012).

353. See supra Section I.B.354. See supra Section II.A.2.355. Fischer, supra note 172, at 796.

2019]

INDIANA LAW JOURNAL

well, although this can be done on a case-by-case basis through briefs by parties, aswell as amici.

Increasing awareness of the sophisticated doctor schema is consistent withexisting concerns. For example, the U.S. Accreditation Council for ContinuingMedical Education has expressed concern that residents learn how promotionalactivities can influence their prescribing information and the World HealthOrganization has urged countries to include information on commercial marketingstrategies in formal training.3 6 In addition, some academic institutions have alsolimited promotional activity,3 7 and some doctors already turn away salespeople.3 8

However, more consistent revelation of marketing influences is important not onlyin medical schools, but also through continuing medical education, since doctors ofall ages are vulnerable to marketing.

Increased awareness can build upon the work of Healthy Skepticism, aninternational organization that aims to better inform doctors about their susceptibilityto subtle techniques used by advertisers.3'

9 They have already created trainingmodules about how people are vulnerable to persuasion,360 and also have training toreduce overconfidence bias, including the mistaken belief by many doctors thatwithout any training, they can easily distinguish misleading claims from justifiedones.36 1 Some of these modules could be easily incorporated into medical education(whether initial or continuing). There have also been promising results for bettereducating doctors and residents to recognize that they are vulnerable to marketing bylimiting their overconfidence.362 For example, one established method is to expose

356. ACCREDITATION COUNCIL FOR CONTINUING GRADUATE MED. EDUC., PRINCIPLES TO

GUIDE THE RELATIONSHIP BETWEEN GRADUATE MEDICAL EDUCATION, INDUSTRY, AND OTHER

FUNDING SOURCES FOR PROGRAMS AND SPONSORING INSTITUTIONS ACCREDITED BY THE

ACGME (2011), https ://www.acgme.org/Portals/O/PFAssets/PublicationsPapers/GME-Funding-Industry-Other-Sources.pdf [https://perma.cc/HZ2C-J6R4]; BARBARA MiNTZES,

WORLD HEALTH ORG. & HEALTH ACTION INT'L, EDUCATIONAL INITIATIVES FOR MEDICAL AND

PHARMACY STUDENTS ABOUT DRUG PROMOTION: AN INTERNATIONAL CROSS-SECTIONAL

SURVEY (2005).357. E.g., David L. Coleman, Establishing Policies for the Relationship Between Industry

and Clinicians: Lessons Learned from Two Academic Health Centers, 83 ACAD. MED. 882(2008); Larkin et al., Physician Prescribing, supra note 64; see also PEW CHARITABLE TR.,

CONFLICT-OF-INTEREST POLICIES FOR ACADEMIC MEDICAL CENTERS 12 (2013)(recommending barring pharmaceutical representatives as one type of best practice).

358. E.g., Robert Lowes, More Physicians Saying 'No Drug Reps Allowed', MEDSCAPE

(Sept. 13, 2016), https://www.medscape.com/viewarticle/868748 [https://perma.cc/CY5G-XSEJ] (noting that approximately thirty-five percent turn away salespeople).

359. See Mansfield, Healthy Skepticism, supra note 132, at 644.360. Id. at 645.361. See id. The organization also aims to disabuse doctors of other unsupported schemas

that the industry promotes, such as the fiction that newer drugs are in general better, in contrastto the reality where only about three percent are major advances. Id.

362. See Peter R. Mansfield, Joel Lexchin, Leana S. Wen, Luisella Grandori, Christopher

P. McCoy, Jerome R. Hoffman, Joana Ramos & Jon N. Jureidini, Educating HealthProfessionals About Drug and Device Promotion: Advocates 'Recommendations, PLOS MED.,Nov. 2006, at 1988 [hereinafer Mansfield et al., Educating Health]; see Sagarin et al., supranote 126.

[Vol. 94:773


individuals to a standard sales technique, allow individuals to express any beliefs,and then debunk them, as well as explain what misleading techniques were used.363

Importantly, this has been shown to reduce overconfidence in feeling "skilled" atcritically appraising information from health representatives.364

In addition, informing doctors of the benefits of limiting industry interaction,coupled with how to substitute the perceived benefits of industry marketing, wouldalso be valuable. For example, doctors frequently state that they find detailing aconvenient way both to learn about drugs and get drug samples for their patients,while also praising the social and informative aspects of lunches paid for bycompanies.365 However, some of these benefits are illusory. One family medicinepractice that took an inventory of free drug samples found that few were "first-line"drugs for most common illnesses, and less expensive alternatives were available fora majority.366 In addition, for a relatively low cost, the practice maintainededucational group lunches that were not influenced by marketing and insteadprovided educational information about both new and old drugs based on peer-reviewed articles.367 Although some doctors were initially resistant and skeptical ofremoving all industry influence, seeing actual data regarding number of visits andsamples was helpful.368

This example illustrates that it may be important not only to make doctors awareof the influence of marketing, but perhaps also to increase awareness with their ownpeers. As noted earlier, doctors can be skeptical of those outside their own circles.Also, there are some doctors, albeit a minority, that do recognize vulnerability tomarketing. Accordingly, there should be a pool of doctors that can assist in increasingawareness among all doctors. In addition, if medical schools can follow the lead ofmany academic centers in not only limiting pharmaceutical interaction, but alsoeducating students about their vulnerability, this could strongly facilitate increasedknowledge for a new generation of doctors.

363. Mansfield et al., Educating Health, supra note 362, at 1989.364. Michael S. Wilkes & Jerome R. Hoffman, An Innovative Approach to Educating

Medical Students About Pharmaceutical Promotion, 76 AcAD. MED. 1271, 1274 (2001).365. E.g., David Evans, Daniel M. Hartung, Denise Beasley & Lyle J. Fagnan, Breaking

Up is Hard to Do: Lessons Learnedfrom a Pharma-Free Practice Transformation, 26 J. AM.BD. FAM. MED. 332, 334 (2013). Even doctors that consider information from representativesto be biased may enjoy detailing visits for the physical relationship or for gifts received fromdoctors. Fischer et al., supra note 172, at 797-98; Spiller & Wymer, supra note 171, at 94(internal citations omitted). In addition, many doctors consider the detailing usefulinformation. See supra note 171 and accompanying text; see also Sorrell v. IMS Health Inc.,564 U.S. 552, 578 (2011) (noting that doctors find detailing "very helpful").

366. Evans et al., supra note 365, at 334.367. Id. at 334, 336 (noting that since some providers were initially resistant to terminating

free samples and all visits by pharmaceutical representatives, intermediate steps were taken totransition providers and ultimately providers could see that there were cheaper optionsavailable).

368. Some noted that they needed industry information to be current and there would benegative social harm, but eventually all agreed that this was a positive step. Id. at 336.

2019]

INDIANA LAW JOURNAL

2. Beyond Conflicts of Interest - Reconsidering Drug Samples and Gifts

Although industry influence resulting in conflicts of interest prompted enactmentof the Sunshine Act, the cognitive biases discussed here provide additional reasonsto be concerned about manipulative marketing not only with off-label uses of drugs,but for all prescription drugs. After all, the industry uses the same types of marketingfor all uses of prescription drugs.

Importantly, whereas the Sunshine Act was promulgated under the theory thatmore transparency would limit industry influence since doctors would be hesitant toaccept money they would need to disclose, this legislation inherently fails to addresskey cognitive biases that still subject doctors to the influence of industry marketing.For example, the Sunshine Act seems to assume that smaller value items, such asdrug samples and gifts under $100 have no impact, and thus exempts these fromdisclosure.369 However, both of these types of items can and do have an impact-even though the law does not currently recognize this and doctors assume that thereare no dangers.37

' The industry invests over $5 billion a year on these practices,which yield profitable outcomes for companies.371 Studies consistently show thatsamples impact prescriptions.372 This makes sense given the cognitive bias ofavailability that impacts everyone, including doctors; since samples are available,doctors are naturally inclined to think about them for prescribing. 373

369. Patient Protection & Affordable Care Act, Pub. L. No. 111-148, §1128G(e)(10)(B)(i)-(ii), 124 Stat. 119, 696 (codified at 42 U.S.C. § 18001 (2012)).

370. See Allan S. Brett, Wayne Burr & Jamaluddin Moloo, Are Gifts from PharmaceuticalCompanies Ethically Problematic?: A Survey of Physicians, 163 ARCHIVE INTERNAL MED.

2213 (2003); Steinman et al., supra note 174. Although some scholars have previouslyrecognized the impact of small gifts to influence doctors, arguments that these resulted inconflicts of interests perhaps led to the improper assumption that they are not of concern. See,e.g., Susan Chimonas, Troyen A. Brennan & David J. Rothman, Physicians and DrugRepresentatives: Exploring the Dynamics of the Relationship, 22 J. GEN. INTERNAL MED. 184(2007); Jason Dana & George Lowenstein, A Social Science Perspective on Gifts to Physiciansfrom Industry, 290 J. AM. MED. ASS'N 252 (2003); Wazana, supra note 20.

371. E.g., Persuading the Prescribers: Pharmaceutical Industry Marketing and ItsInfluence on Physicians and Patients, PEW (Nov. 11, 2013), https://www.pewtrusts.org/en/research-and-analysis/fact-sheets/2013/1 1/1 l/persuading-the-prescribers-pharmaceutical-industry-marketing-and-its -influence -on-physicians -and-patients [https ://perma.cc/APM7-B6G5].

372. E.g., John M. Boltri, Elizabeth R. Gordon & Robert L. Vogel, Effect ofAntihypertensive Samples on Physician Prescribing Patterns, 34 FAMILY MED. 729 (2002);Lisa D. Chew, Theresa S. O'Young, Thomas K. Hazlet, Katharine A. Bradley, CharlesMaynard & Daniel S. Lessler, A Physician Survey of the Effect of Drug Sample Availabilityon Physicians'Behavior, 15 J. GEN. INTERNAL MED. 478 (2000); Fickweiler et al., supra note172, at 3; Adriane Fugh-Berman & Shahram Ahari, Following the Script. How Drug RepsMake Friends and Influence Doctors, PLOS MED., Apr. 2007, at 621, 624.

373. In addition, direct to consumer advertising may prompt consumers to specificallyrequest brand name drugs. Richard L. Kravitz, Ronald M. Epstein, Mitchell D. Feldman, CarolE. Franz, Rachman Azari, Michael S. Wilkes, Ladson Hinton & Peter Franks, Influence ofPatients' Requests for Direct-to-Consumer Advertised Antidepressants: A RandomizedControlled Trial, 293 J. AM. MED. ASS'N 1995, 1998-99 (2005) (reporting that for some typesof issues, doctors were more likely to prescribe the specific drug requested by a patient) as a

[Vol. 94:773


Doctors who embrace the sophisticated doctor schema may believe that they areinvulnerable to influence from these low-budget items and also provide justificationsthat are not consistent with reality. For example, doctors often report that samplesmake patients happy, or result in lower costs especially for low-income patients,74

and that receiving them has no impact on their prescriptions."' However, in fact,data show that drug samples are typically provided to high, rather than low-incomepatients and also ironically lead to higher costs, since (1) typically a patient will needmore than the initial sample and (2) samples are inevitably of high-priced patenteddrugs.76 Furthermore, studies indicate that patients actually disapprove of bothsamples and gifts.3

Given this reality, the question is what the next step should be. An easy first stepwould be informing doctors of these data more broadly, perhaps through continuingmedical education. In particular, informing doctors of colleagues who havesuccessfully taken this step and have come to see the benefits, including those whowere initially skeptical, might be helpful.3"8 One small practice group has provideda template for how a group can voluntarily extricate itself from such pharmaceuticalmarketing.37 9 Some doctors may still be resistant to believing the data, given theirpreexisting biases. Accordingly, a more drastic step would be to legally barcompanies from providing drug samples. Of course, this is a major change that would

result of advertising); Michael S.Wilkes, Robert A. Bell & Richard L. Kravitz, Direct-to-Consumer Prescription Drug Advertising: Trends, Impact, and Implications, 19 HEALTH AFF.110, 118 (2000) (finding that in one survey, nineteen percent of people asked for a specificallyadvertised drug). When consumers ask for such drugs, doctors likely have samples sincecompanies use both methods to market drugs. Gagnon & Lexchin, supra note 53, Table 1(noting expenditures on detailing as well as advertising to consumers).

374. E.g., Fischer et al., supra note 172, at 797 (noting that doctors claimed that samplesmake patients happy and even that patients expect samples); Tim Lahey, The High Costs of"Free" Drug Samples, CLINICAL & TRANSLATIONAL GASTROENTEROLOGY, 2014, at 1 (notingthat some doctors believe samples benefit low-income patients).

375. See, e.g., Richard F. Adair & Leah R. Holmgren, Do Drug Samples Influence ResidentPrescribing Behavior? A Randomized Trial, 118 AM. J. MED. 881, 883 (2005); Brian Hodges,Interactions with the Pharmaceutical Industry: Experiences and Attitudes of PsychiatryResidents, Interns and Clerks, 153 CAN. MED. ASS'N J. 553, 558 (1995); Dana Katz, Arthur L.Caplan & Jon F. Merz, All Gifts Large and Small: Toward an Understanding of the Ethics ofPharmaceutical Industry Gift-Giving, 10 AM. J. BIOETHICS 11, 12 (2010); Steinman et al.,supra note 174, at 555.

376. See Sarah L. Cutrona, Steffie Woolhandler, Karen E. Lasser, David H. Bor, DannyMcCormick & David U. Himmelsteir, Characteristics of Recipients of Free PrescriptionDrug Samples: A Nationally Representative Analysis, 98 AM. J. PUB. HEALTH 284, 285-87(2008); Lahey, supra note 374, at 1.

377. Lynn Eaton, Readers Want Transparency in Link Between Doctors and Drug Firms,326 BRIT. MED. J. 1352, 1352 (2003); Robert V. Gibbons, Frank J. Landry, Denise L. Blouch,David L. Jones, Frederick K. Williams, Catherine R. Lucey & Kurt Kroenke, A Comparisonof Physicians' and Patients' Attitudes Toward Pharmaceutical Industry Gifts, 13 J. GEN.

INTERNAL MED. 151, 152-53 (1998); James Jastifer & Sarah Roberts, Patients 'Awareness ofand Attitudes Toward Gifts from Pharmaceutical Companies to Physicians, 39 INT'L J.HEALTH SERVS. 405, 411 (2009).

378. Fischer et al., supra note 172, at 798.379. See supra notes 365-368 and accompanying text.

2019]

INDIANA LAW JOURNAL

likely be vigorously opposed by the well-funded pharmaceutical industry that spendsbillions a year on promotional activity, with the vast majority being spent on detailingand drug samples, such that it may not be a realistic recommendation in the nearfuture. Nonetheless, doing so is consistent with cognitive bias studies.38

3. Other Mechanisms to Minimize Public Hann From Schemas

Given the serious potential public health hazards associated with off-label use inaddition to the lack of awareness of schemas, additional action could be taken tominimize harm, including mechanisms that would require no change to the FirstAmendment law or challenging existing schemas held by doctors. In particular, off-label use that is not medically supported could, and arguably should, be limited. Thiswould be consistent with an evidence-based approach to medicine that many suggestis how medicine should be practiced, even if that does not always happen. Theexisting law governing Medicare payments could be amended to limit payments foroff-label use unless it is supported by high-quality evidence and what counts asevidence is revisited. After all, current guidelines are fairly permissive and can relyon compendiums38 1 that studies have shown to be sometimes not accurate or up todate,382 as well as medical articles that are not always reliable.38 3 Of course, thiswould likely be strongly opposed by the pharmaceutical industry. Alternatively,private insurance companies could endeavor to modify their reimbursement forunsupported off-label use. Insurance companies have a clear interest in not payingfor unnecessary treatment and in recent years have been more strictly limitingpayment for drugs with various tiered formularies. The proposed change would bean expansion of the existing approach to not only consider drugs within the sameclass differently, but also the use of the drug. This may be complicated under theexisting system where doctors generally do not need to indicate the use for a drug ona prescription. However, electronic medical records make this easier, indicating thatthis is not impossible.384

380. Gagnon & Lexchin, supra note 53, at 29, 31.381. Medicare covers all "medically accepted indications," which includes not just FDA-

approved indications, but also compendia-based information. 42 U.S.C. § 1396r-8(k)(6)(2012) (defining medically accepted indication for Medicaid to include certain compendia);see also 42 U.S.C. § 1396r-8(g)(1)(B)(i) (2012); 42 C.F.R. § 423.100 (2015) (definingMedicare Part D drug with reference to Medicaid definition of medically accepted indication).

382. E.g., Amy P. Abernethy, Gowri Raman, Ethan M. Balk, Julia M. Hammond, Lori A.Orlando, Jane L. Wheeler, Joseph Lau & Douglas C. McCrory, Systematic Review: Reliabilityof Compendia Methods for Off-Label Oncology Indications, 150 ANNALS INTERNAL MED. 336,341-42 (2009) (finding that compendia do not always link recommendations to evidence andoften are not the most current evidence used).

383. Medical literature may also be considered for Medicaid reimbursement, but is notalone adequate except for anticancer treatment. 42 U.S.C. § 1396r-8(g)(1)(B)(ii) (2012); 70Fed. Reg. 4194, 4228-29 (Jan. 28, 2005) (to be codified at 42 C.F.R. pt. 423); see also JenniferL. Herbst, How Medicare Part D, Medicaid, Electronic Prescribing, and ICD-I0 CouldImprove Public Health (But Only if CMS Lets Them), 24 HEALTH MATRIX: J. L.-MED. 209,216 (2014).

384. See Eguale et al., supra note 96, at 56 (noting that in Quebec documentation oftreatment indication is mandatory).

[Vol. 94:773


4. Towards More Independent and Accessible Information

A major issue with pharmaceutical marketing is its widespread availability andaccessibility, making doctors more likely to rely on it even if they realize it is skewedand not well supported. Even informing doctors that they are relying on theinformation unduly is not likely to be ideal since studies repeatedly show that allindividuals tend to rely on easily accessible information. To counteract industrymarketing, not only do different sources of information need to exist, but they needto be as available and accessible as existing marketing. If such independent andaccessible information exists when drugs are first introduced, there would be less ofa need to train doctors about marketing influences. But, since such independentinformation currently does not exist, greater awareness is essential, even if the bestpossible results from that awareness are still suboptimal. Importantly, if the majority,rather than a minority, of doctors were aware of their vulnerability to marketing,perhaps that could help to better protect patients since doctors do consult with peersand seem to generally value peer opinions, as opposed to those outside theirprofession.

There is already some recognition of the need to develop independent data in thebroader medical context. A 2007 study found that less than half of recommendedtreatments are based on sound science.38 Since then, Congress has includedappropriations to fund comparative effectiveness studies.386 These studies areessential to complement initial studies done by self-interested drug companies thattypically only evaluate a proposed new drug versus a placebo, rather than existingtreatments, and also under experimental, rather than typical scenarios. Even in thelimited situations where companies compare their drugs to others, the study may beskewed due to different dosages or other modifications. Such maneuvers shouldarguably be easy for physicians to spot. And, if companies presented all informationto doctors, that might be true. However, there is no incentive for companies to do so.

Assuming independent data can be developed, the next step is to make the dataeasily available and accessible to doctors. One way to do so is to actually mirror theeffective drug detailing done by companies with "academic detailing," wherebyscientifically trained individuals with no profit-based agenda are providinginformation in a one-on-one format that has been shown to be accessible andconvenient for doctors.38 7 Studies have shown some success in improving care

385. INST. OF MED. OF THE NAT'L AcADs., LEARNING WHAT WORKS BEST: THE NATION'S

NEED FOR EVIDENCE ON COMPARATIVE EFFECTIVENESS IN HEALTH CARE 78 (2011) (citing 2007study).

386. American Recovery and Reinvestment Act, Pub. L. No. 111-5, 123 Stat. 115 (2009);see also Robert Pear, U.S. to Compare Medical Treatments, N.Y. TIMES (Feb. 15, 2009),https://www.nytimes.com/2009/02/16/health/policy/16health.html [https ://perma.cc/72L7-PAG2].

387. See Jerry Avorn, Academic Detailing: "Marketing" the Best Evidence to Clinicians,317 J. AM. MED. ASS'N 361 (2017); Michael A. Fischer & Jerry Avorn, Academic DetailingCan Play a Key Role in Assessing and Implementing Comparative Effectiveness ResearchFindings, 31 HEALTH AFF. 2206 (2012). In addition, although academic detailing is likely themost accessible method, even continuing medical education can be made more effective if thetraditional pure lecture styles are replaced by more interactive formats that have been found

2019]

INDIANA LAW JOURNAL

through academic detailing. 311 Of course, although academic detailing is beneficial,it would be tough to rely on academic detailing alone to completely counteractmisleading marketing messages, contrary to the presumptions of the Sorrell majorityopinion. After all, the pharmaceutical industry has substantial funds to promote itsown products. Most academic detailing focuses on one or two limited areas.Nonetheless, greater attention to academic detailing is better than none at all. It isalso a strategy that has no commercial speech problems, such that some scholarsrecently suggested it as a strategy to address Sorrell.389

5. Additional Support for Reconsidering the Drug Development Process

The data presented here also provides additional support for questioning theoverall system of drug development, and not simply issues concerning off-label useof drugs. This Article shows that companies have an incentive to promote off-labeluses that may not be supported by evidence. However, the problems with drugdevelopment and marketing are much more extensive. Not only do companies havean incentive to market off-label uses during the limited term of patent protection tomaximize profits, but they also have an incentive to develop the most profitabledrugs, which are not necessarily the ones that are most socially desirable.39

' Forexample, most drugs are developed for relatively wealthy countries that have theability to pay.391 This is not a new discovery. Indeed, legislation exists to try toencourage companies to promote drugs that impact smaller classes with a variety ofincentives that include tax advantages, as well as commercial exclusivity.392

useful in a wide array of educational contexts.388. E.g., Avom, supra note 387 (noting that since research started in the 1980s, studies

have shown success in improving care in a variety of settings, including controlling use ofsedating medications in nursing homes and reducing overuse of antibiotics).

389. George R. Gooch, J. James Rohack & Marisa Finley, The Moral From Sorrell:Educate, Don't Legislate, 23 HEALTH MATRIX 237, 262 (2013).

390. E.g., Steven J. Hoffman & Karen So, Assessing 15 Proposals for PromotingInnovation andAccess to Medicines Globally, 80 ANNALS GLOBAL HEALTH 432, 433 (2014).Some have suggested that since the patent system contributes to this problem, alternatives areneeded to encourage socially valuable innovation. E.g., Ho, supra note 11, at 367-71(discussing proposals such as a health impact fund, publicly funded clinical trials, and a drugdevelopment corporation); UNITED NATIONS, REPORT OF THE UNITED NATIONS SECRETARY-

GENERAL'S HIGH-LEVEL PANEL ON ACCESS TO MEDICINES: PROMOTING INNOVATION AND

ACCESS TO HEALTH TECHNOLOGIES 7 (2016).391. This is referred to as the 10/90 gap: less than ten percent of resources are devoted to

the health needs of developing countries where over ninety percent of preventable deathsoccur. Roderik F. Viergever, The Mismatch Between the Health Research and Development(R&D) that Is Needed and the R&D that is Undertaken: An Overview of the Problem, theCauses, and Solutions, 6 GLOBAL HEALTH ACTION 22450, 22450 (2013). This is such asubstantial problem that it is included in a "scorecard" of how well pharmaceutical companiesare acting in the interest of poorer countries. ACCESS TO MED. FOUND., ACCESS TO MEDICINE

INDEX 2018: METHODOLOGY REPORT (2018) (explaining basis for scoring that includes notonly research into needs of developing countries, but also providing access to neededmedications).

392. Orphan Drug Act, Pub. L. No. 97-414, 96 Stat. 2049 (1982) (codified as amended at

[Vol. 94:773


However, the industry has managed to game this well-intended legislation to obtainhandsome profits while sometimes also charging exorbitant prices to consumers.393

In light of these issues, there have been proposals to overhaul the domestic andinternational system.394 Admittedly, a major overhaul to the drug developmentprocess would be a major change and a complete discussion is beyond the scope of

21 U.S.C. § 360aa-360ee (2012), 26 U.S.C. § 45C (2012), 42 U.S.C. § 236 (2012)); CouncilRegulation 141/2000, 2000 O.J. (L 18) 1.

393. Anticompetitive Abuse of the Orphan Drug Act: Invitation to High Prices: HearingBefore the Subcomm. on Antitrust, Monopolies & Business Rights of the S. Comm. on theJudiciary, 102d Cong. 198 (1992) (statement of JohnL. Castello); Jess G. Thoene, Curing theOrphan DrugAct, 251 Sci. 1158, 1158 (1991); Dina Gusovsky, How a Blockbuster Drug CanBecome a Subsidized 'Orphan', CNBC (Dec. 2, 2015, 11:27 AM), https://www.cnbc.com/2015/12/u1/an-obscure-fda-rule-adding-to-drug-company-profits.html [https:/perma.cc/3DCX-GSZW]; Sarah Jane Tribble & Sydney Lupkin, Drugs for Rare DiseasesHave Become Uncommonly Rich Monopolies, NPR (Jan. 17, 2017, 4:59 AM), https://www.npr.org/sections/health-shots/2017/01/17/509506836/drugs-for-rare-diseases-have-become-uncommonly-rich-monopolies [https://perma.cc/HTQ5-CZEA]. Indeed, drugsoriginally approved as orphan drugs have high potential for off-label marketing to expandcompany profitability. E.g., Marc-Andrd Gagnon, New Drug Pricing: Does It Make AnySense?, 24 PRESCRWE INT'L 192 (2015); Shannon Gibson, Hamid R. Raziee & TrudoLemmens, Why the Shift? Taking a Closer Look at the Growing Interest in Niche Markets andPersonalized Medicine, 7 WORLD MED. & HEALTH POL'Y 3, 15 (2015). There are someproposals to modify the system. E.g., Sarah Jane Tribble, House Republicans Aim to Yank TaxCreditsfor Orphan Drugs, KEN (Nov. 2, 2017), https://khn.org/news/house-republicans-aim-to-yank-tax-credits-for-orphan-drugs [https ://perma.cc/RZ3K-WFGL].

394. E.g., FINKELSTEIN & TEMIN, supra note 13; AtDAN HOLLIS & THOMAS POGGE, THE

HEALTH IMPACT FUND: MAKING NEW MEDICINES ACCESSIBLE FOR ALL (2008); UNITEDNATIONS, REPORT OF THE UNITED NATIONS SECRETARY-GENERAL'S HIGH-LEVEL PANEL ONACCESS TO MEDICINES: PROMOTING INNOVATION AND ACCESS TO HEALTH TECHNOLOGIES

(2016); Tim Hubbard & James Love, A New Trade Framework for Global Healthcare R&D,PLoS BIOLOGY, 2004, at 0147. Even relatively modest suggestions, such as changing the lawto permit Medicare to reduce prices attract controversy. Medicare Negotiation andCompetitive Licensing Act, H.R. 6505, 115th Cong. (2018); Peter Rheinstein, Gutting DrugPatents Puts Lives at Risk, NAT'L REV. (Oct. 2, 2018, 6:30 AM), https://www.nationalreview. com/20 18/1O/gutting-drug-patents-puts-lives-at-risk [https ://perma.cc/AZ83-CFRR]. There are also proposals for drug cost transparency laws that, while not reformingthe entire system, aims to at least require accountability; although no federal legislation haspassed, some states have been successful. Prescription Drug Price Transparency Act, H.R.1316, 115th Cong. (2017); Ameet Sarpatwari, Jerry Avom & Aaron S. Kesselheim, StateInitiatives to Control Medication Costs Can Transparency Legislation Help?, 374 NEw ENG.J. MED. 2301 (2016); Zachary Brennan, Oregon Governor Signs Drug Price TransparencyBill, REG. AFF. PROF. SoC'Y (Mar. 13, 2018), https://www.raps.org/news-and-articles/news-articles/20 18/3/oregon-bill-on-drug-price-transparency-headed-to-g [https ://perma.cc/3BZK-764V]; Ron Lanton, III, The Current State of Drug Price Legislation,SPECIALTY PHARMACY TIMES (Nov. 16, 2017), https://www.specialtypharmacytimes.com/publications/specialty-pharnacy-times/20 17/november-20 17/the-current-state-of-drug-price-legislation [https://perma.cc/RQP2-8GBV]; Ed Silverman, Vermont Becomes FirstState to Require Drug Makers to Justify Price Hikes, STAT (June 6, 2016), https://www. statnews.com/pharmalot/2016/06/06/vermont-drug-prices-transparency [https://perma.cc/GG2Z-A54K].

2019]

INDIANA LAW JOURNAL

this Article. However, the data in this Article lends further support to arguments thatthe current system is broken, and more serious attention to fixing it is required.

CONCLUSION

The intersection of commercial speech and FDA regulation of unapproved usesof FDA-approved drugs is an important area to commercial drug companies, as wellas policy makers concerned about public health. Given recent judicial expansion ofcommercial speech, as well as pressure on the FDA to further liberalize corporatespeech concerning uses without clear scientific foundation, it is important to considerthe appropriate balance. This Article has revealed, and debunked, two key schemasprevalent in key cases, as well as among doctors and policy makers. A betterunderstanding of these schemas would inform appropriate changes to commercialspeech law to mirror reality. Moreover, these schemas also support more systematicchanges to drug development and marketing in favor of an independent andevidence-based system. In addition, the existence of these schemas provide strongsupport for issues beyond off-label promotion of drugs. In particular, the schemassuggest that the current minimal scrutiny of dietary supplements with a disclaimermay not be wellfounded. Considering that doctors are more sophisticated thanconsumers and still vulnerable to commercial advertising, the schemas revealed herealso suggest that direct-to-consumer advertisement should be more, rather than less,restricted. Although it is unlikely that all of these changes could be made, evenunderstanding the existence and operation of these schemas is an important first steptowards an improved understanding that should yield the informed results that allagree are desirable.

[Vol. 94:773